Stereo Vision Based Advanced Driver Assistance System

Size: px
Start display at page:

Download "Stereo Vision Based Advanced Driver Assistance System"

Transcription

1 Stereo Vision Based Advanced Driver Assistance System Ho Gi Jung, Yun Hee Lee, Dong Suk Kim, Pal Joo Yoon MANDO Corp ,Gomae-Ri, Yongin-Si, Kyongi-Do, , Korea Phone: (82) Fax: (82) This paper describes a stereo vision based obstacle detection algorithm, which is the core component of advanced driver assistance system incorporating lane departure warning, forward collision warning and avoidance. The proposed vision system recognizes the road lane, on which host vehicle is traveling, by template matching on the bird s eye view of forward scene. The recognition of road lane uses an assumption that a lane marking is a pair of neighboring rising and falling edge and a road lane is a pair of lane marking with a fixed distance. ROI (Region Of Interest) is established according to the recognized ego-lane because preceding vehicle on the ego-lane is expected to be a potential threat to host vehicle. After the establishment of ROI, vision system generates disparity histogram by feature based stereo matching. Because the preceding vehicle has a large amount of vertical edges with the same disparity, it forms a peak in the disparity histogram. Consequently, the preceding vehicle can be detectable by simple thresholding. The threshold of peak detection is designed to vary with respect to disparity, i.e. distance, considering the fact that obstacle appears smaller as its distance becomes further. Detected peaks are verified by the comparison of edge and color between left and right image. Ego-lane based ROI establishment and feature based stereo matching drastically reduce computational burden. Furthermore, disparity histogram based obstacle detection is proved to be robust because it captures big picture successfully ignoring the details. The effect of ego-lane based ROI and adaptive thresholding is verified by experiments with real vehicle. 1. INTRODUCTION Recently, there seems to be enormous needs of intelligent vehicle [1]. Economically practical solutions from fast evolving electronics and sensor industries are expected to realize what has been thought to be impossible in the near future. Introduction of environmental sensor such as radar, vision and GPS allows advanced driver assistance system to recognize surrounding situation. Furthermore, intelligent control algorithm utilizing recognized environment enables intelligent vehicle to automatically enhance the convenience and safety of driver. Among several kinds of environmental sensor, vision sensor is taking intensive attention because of its compatibility with human visual system. In other words, because current driving environment is designed for human without the consideration of automation, especially depending on human visual perception capability, vision sensor is thought to be ideal to interpret a driving situation. It can be easily approved by the plentifulness of recent vision applications such as traffic surveillance [2], license plate / traffic sign recognition [3, 9], free parking site localization [4], lane detection [5, 6] and obstacle / pedestrian detection [6-8]. Stereo vision system has advantages over range sensor system including laser radar and mm-wave radar: it can recognize visual cues such as road lane, traffic sign and detailed object shape. Although some companies are trying to develop monocular vision system that can recognize road lane and obstacles [7], it is certain that stereo vision is more intuitive than monocular vision for the recognition of 3D information. Fig. 1 shows the advantage of stereo vision over monocular vision. In monocular vision, points on a certain sight line are mapped onto one point. However, stereo vision can distinguish these points using displacement between left and right image. This paper explains our approach to stereo vision based collision avoidance system incorporating several vision based functions such as lane departure warning, collision warning and avoidance. X X X Fig. 1 Monocular camera vs. stereo camera 2. SYSTEM ARCHITECTURE The advanced driver assistance system proposed in this paper consists of 5 main components: stereo vision based obstacle distance measurement, dynamic model of ego-vehicle, collision avoidance algorithm, active x L x L X x R X x R

2 braking and HMI (Human Machine Interface). The dynamic model of ego-vehicle estimates required state variable of ego-vehicle utilizing the existing sensors already installed on the vehicle such as wheel speed, lateral acceleration, yaw rate and steering angle. The collision avoidance algorithm considers current vehicle state and obstacle state, and then sends required braking command to active braking system via CAN in order to avoid upcoming collision. The active braking system should be able to generate required braking force without driver s pedal maneuver. In the proposed system, the Mando MGH-Series ESP is used as an active braking actuator. Fig. 2 shows the brief architecture of the system. The stereo vision based obstacle distance measurement, the main focus of this paper, consists of 4 phases: ego-lane based ROI (Region of Interest) establishment, feature based stereo matching, peak detection by adaptive threshold in disparity histogram and verification of candidate peaks by edge/color similarity. The proposed stereo matching measures the distance of obstacle, i.e., preceding vehicle, using disparity histogram. It assumes that preceding vehicle will make a definite peak in the disparity histogram and peak detection can recognize the existence of the object. To make the disparity histogram emphasize the preceding vehicle and ignore potential disturbances effectively, vision system establishes ego-lane based ROI. To compensate many kinds of variation, adaptive threshold is used in the peak detection. Peak detection in disparity histogram does not need the recognition of object boundary, which may be the major challenge of object recognition system because of its complexity and high computational load. This paper shows that the effect of ego-lane based ROI and adaptive thresholding is proved by experimental comparisons. Finally, the proposed method is compared with laser radar in the aspect of detection range and response time. 3. ROI ESTABLISHMENT ROI (Region Of Interest) for the following operations is established according to recognized ego-lane. ROI establishment is important in two aspects: to reduce computational load and to improve the robustness of system by ignoring useless image portion. Furthermore, because the established ROI contains sufficient image portion to detect the distance of preceding vehicle, explicit object boundary detection can be omitted. In this paper, a template matching on the bird s eye view of forward scene is used for the detection of ego-lane. 3.1 The Bird s Eye View of Forward Scene The bird s eye view of an image is a virtual image taken from the sky as if a flying bird sees the same scene. In general, bird s eye view is introduced to eliminate or mitigate the effect of perspective distortion. When a man sees a scene with his own eye or camera, an object located further appears smaller than the nearer in spite of the same length. This kind of distortion is common in pinhole camera model and called perspective distortion. Fig. 3(a) shows the perspective distortion. Although the lane in front of host vehicle is expected to have a fixed width, the width decreases as the distance becomes further. In contrast to the captured image, the bird s eye view of the image, shown in Fig. 3(b), reveals the fixed width of road lane. Furthermore, lane marking also has a fixed width in the bird s eye view. However, it is noticeable that the bird s eye view of an image considers every pixel to be attached onto a plane surface and can not help a strange distortion caused by an object above the ground surface such as preceding vehicle. (a) Captured scene (b) Bird s eye view Fig. 3. The bird s eye view of a forward scene The bird s eye view of an image is generated with homography matrix H. Homography defines the one-to-one relation between pixels of two images, in this case, between captured image pixel and bird s eye view pixel. If X is the coordinate of image pixel and X' is the coordinate of corresponding bird s eye view pixel, homography matrix H interconnects them like equation (1). X, X' and H uses homogeneous coordinate system. X'= HX (1) Fig. 2 System block diagram Because X-X' pairs are available from test image and desired bird s eye view, homography matrix H can be estimated by SVD (Singular Value Decomposition). H generally has an inverse matrix H -1 corresponding to

3 the inverse mapping. 3.2 Ego-Lane Recognition A template matching technique is used to detect the lane marking on bird s eye view image. The template is implemented under the assumption that a lane marking is a neighboring falling and rising edge pair with a fixed distance w in the horizontal direction [6]. Equation (2) shows how the template matching result image T(x,y) is calculated. In this equation, H and W are the height and width of the bird s eye view, respectively. B(x,y) designates the intensity of a pixel (x,y). Consequently, T(x,y) measures how the pixel (x,y) is likely to be lane marking by examining the intensity differences between the pixel and left/right pixels respectively with distant ±w. Binarization of T(x,y) by applying thresholding generates a lane marking candidate image, as shown in Fig. 4(a). H W T ( x, y) = (2B( x, y) B( x w, y) B( x + w, y)) (2) y= 0 x= 0 Detecting lane marking pairs with a fixed distance in the horizontal direction recognizes road lanes in image row y. Because the ego-lane is only the image of interest, one road lane locating near to the vertical centerline is selected as an ego-lane from the image row y. Ego-lanes of each image row are expected to totally form ego-lane in the sense of LS (Least Squared). Therefore, curve fitting of ego-lanes of each image row can be a good estimate of ego-lane. Fig. 4(b) shows the result of ego-lane recognition and the central dotted line is the centerline of ego-lane. perspective view. ROI is established as an image portion under two curves, which is the recognized ego-lane boundaries moved upward with a fixed offset. 4. STEREO MATCHING In this paper, feature based stereo matching finds out the distance information of image pixels. In the case of automotive vision, it is known that vertical edges are sufficient to detect noticeable objects []. Consequently, stereo matching using only the vertical edges can drastically reduce the computational load. The feature based stereo matching consists of pixel classification and similarity based matching. 4.1 Feature Detection: Pixel Classification Pixel classification investigates the intensity differences between a pixel and 4 directly connected neighbors so as to assign the pixel class reflecting the intensity configuration. It is known that the feature based stereo matching with pixel class is fast and robust to noise []. Equation (3) shows how the relationship between g(x) and g(i) is encoded. g(i) is the gray value of neighboring pixel and Fig. 6(a) shows how the pixel index i is defined. Consecutively, equation (4) assigns a class to the investigated pixel by truncating 4 codes like Fig. 6(b). It is obvious that a pixel of smooth surface will be classified as zero class and a pixel of edge will be classified as non-zero class. To reduce the effect of threshold T, histogram equalization or adaptive threshold can be used. (a) Neighboring pixel index (b) Class encoding Fig. 6 Neighboring pixel indexing and class encoding (a) Lane marking candidates (b) Recognized ego-lane Fig. 4 Lane marking and ego-lane recognition result 3.3 ROI Setting 1 if g( i) g( x) > + T d( i) = 2 if g( i) g( x) < T (3) 0 else where, g(.) means the gray value 3 i d ( x) = 3 d( i) (4) i= 0 (a) Recognized ego-lane (b) Established ROI Fig. 5 ROI establishment based on recognized ego-lane ROI for following operations is established using the recognized ego-lane. Ego-lane on bird s eye view is transformed into ego-lane on perspective view by homography matrix. Fig. 5(a) shows the ego-lane on (a) Original image (b) Classification result Fig. 7 Feature detection result

4 Figure 7(a) shows original image and figure 7(b) shows feature detection result. 6.4% of total pixels are classified as non-zero class. 4.2 Stereo Matching Stereo matching is performed only on pixels classified as vertical edge. Furthermore, stereo matching is composed of step-by-step test sequences through class comparison, class similarity, color similarity and maximum similarity detection. Only correspondence candidates passing the previous test step will be investigated in the next test step S class ( x, ys, ) = f ( Lclass ( x+ u, y+ v), Rclass ( x+ u+ s, y+ v)) (5) 3 3u= 1 v= 1 where, 0, a = b f ( a, b) = 1, a b In general, object attached to the ground surface contributes to a wide range of disparities because it exists in a wide range of distance from host camera. In contrast to the case, object above the ground surface including preceding vehicle tends to contribute to a narrow range of disparities because it exists in a narrow range of distance from the camera. In other words, host camera can observe the backend of preceding vehicle, which is generally a plane parallel to the image plane of the camera. Therefore, vertical edge caused by the preceding vehicle s backend plane is supposed to exist at the same distance. Consequently, preceding vehicle reveals its existence in disparity histogram as a definite peak. []. Fig. 8 is the example of disparity histogram. occurrence S color 1 ( x, y, s) = u= 2v= 2 ( X left ( x + u, y + v) X 5 5 right ( x+ u + s, y + v)) 2 (6) where, X : [R, G, B] S( x, y, s) S ( x, y, s) S ( x, y, s) = (7) color class Assuming that the vertical alignment of stereo rig is correct, the search range of a pixel is limited to a horizontal line with a fixed displacement. First, correspondence test is performed on pixels with the same class as the investigated pixel. Class similarity defined by equation (5) is the measure of how the candidate pixel is similar to the investigated pixel in the sense of 3x3 class window. Color similarity defined by equation (6) is the measure of how the candidate pixel is similar to the investigated pixel in the sense of 5x5 color window. Total similarity defined by equation (7) is the product of the class similarity and the color similarity. If highest total similarity is lower than a certain threshold, the investigated pixel fails to find corresponding point and is ignored. The pixel with highest total similarity is corresponding point. 5. DETECTION OF OBSTACLE DISTANCE The proposed system detects the distance of preceding vehicle by the adaptive thresholding of disparity histogram and verification with respect to disparity. 5.1 Disparity Histogram Preceding vehicle is supposed to form a peak in disparity histogram []. Disparity histogram measures how many pixels have a certain disparity and is implemented as an accumulator array. While system scans whole disparity map, which is the result of stereo matching, each bin of the accumulator array counts up when a pixel has corresponding disparity value. Fig. 8 Disparity histogram disparity 5.2 Adaptive Threshold Adaptive threshold compensates the variation of peak height, which is supposed to be proportional to disparity value. Furthermore, adaptive threshold is designed to vary dynamically reflecting the characteristics of current scene. Although preceding vehicle certainly makes a peak in disparity histogram, the height of peak varies with respect to its disparity value. It can be naturally derived from the fact that distant object appears small and near object appears big. Because near object appears big, the probability of the occurrence of vertical edge pixel is high. Near distance means large disparity. Therefore, near object is expected to generate a peak with high height value at large disparity value. Inversely, because distant object appears small, the probability of the occurrence of vertical edge pixel is low. Distant object is expected to generate a peak with low height value at small disparity value. Constance threshold of the peak detection cannot reflect the disparity-peak height relation. If constance threshold is too low, false detection rate at large disparity value increases. If threshold is set too high to avoid the false detection, preceding vehicle, which is our target to detect, might be missed. Threshold line with respect to disparity is proved to overcome the drawback of constance threshold. Threshold line is a line passing through the origin and its slope is changed according to the characteristics of current scene. There are many factors that influence the occurrence of vertical edge on the vehicle s backend plane. The distance factor is already compensated by the introduction of threshold line. Other factors include

5 illumination condition, reflectance of the vehicle s surface, shape of the vehicle. Continuous length of disparity range over the threshold line is the bottom line of a peak. Long bottom line means too low threshold and short bottom line means too high threshold. Of course, few existence of peak also means the possibility of excessive threshold. Therefore, adjusting the slope of threshold line makes the peak detection adaptive to the current situation. Figure 9 depicts the concept of adaptive threshold. bottom line if too low if too high Fig. 9 Adaptive threshold threshold line 5.3 Verification Sometimes, the peak detection finds several peaks above the threshold line. These peaks are candidates of preceding vehicle. Among the candidates, best one is selected by edge similarity and color similarity. Because correct disparity is the displacement between left and right pixels belonging to the preceding vehicle, right image shifted by the correct disparity will be exactly overlapped on the left image in the area of the preceding vehicle. Edge similarity of a certain disparity d is defined in equation (8) and measures how much portion of edge pixels are overlapped if shifted by the disparity. Edge similarity is normalized measure to have a value between 0 and 1. Color similarity of a certain disparity d is defined in equation (9) and measures how much portion of edge pixels is similar in the sense of R, G, B color code value. Color similarity is also normalized measure to have a value between 0 and 1. Total similarity of a certain disparity d is defined as the product of the color similarity and the edge similarity like equation (). disparity d is able to select the correct disparity. Because disparity histogram might be thought of a kind of dimension reducing transformation, impatient determination of the best peak will make the false detection rate higher. Such a consideration justifies the acceptance of multiple candidates and verification based on the original pixel information. 6. EXPERIMENTAL RESULT Proposed system is validated by in-vehicle tests. Specially, proposed ROI establishment based on ego-lane boundary is validated by comparison with fixed ROI case. Adaptive thresholding is also validated by comparison with constant thresholding. Finally, comparing the distance sequence of cutting-in vehicle with the distance sequence measured by laser radar, I will discuss the advantage of proposed method over laser radar. 6.1 Experimental Setting Stereo camera used in experiments is made with two off the shelf CMOS cameras and its baseline is cm in order to be able to detect distant objects [11]. Figure (a) shows stereo camera module installed at the windshield of test vehicle. Caltech Matlab toolbox is used for stereo camera calibration and rectification [12]. Images used for tests and evaluations are acquired at highway. During the experiments, one laser radar is installed at the front end of test vehicle to record distance measurements, which is used in the validation and performance comparison. (a) Stereo camera installed on the vehicle H W 1 SE ( d) = f ( E ( x, y), E ( x + d, y)) (8) edge left right N y = 0 x = 0 where, 0, a = b f ( a, b) =, E : edge image, 1, a b N: the number of edges I (, xy) I ( x+ dy, ) H W left right 1 RGB,, color ( ) = left (, )(1 ) N y = 0 x = 0 256*3 SE d E x y where, I(x,y)=[R,G,B] at point (x,y) (9) SE( d) = SE ( d) SE ( d) () edge color Correct disparity is expected to have maximum value in the similarity evaluation. It means how much left and right pixels can be overlapped with a certain (b) Laser radar installed on the vehicle Fig. Forward looking sensors on the vehicle 6.2 Effect of Ego-Lane based ROI Distance measurements of an acquired image sequence are calculated by two methods: object detection with fixed ROI and object detection with

6 ego-lane based ROI. Fig. 11(a) is the distance measurements calculated with the fixed ROI method. It is observed that measured distances have large noise. Fig. 11(b) is the distance measurements calculated with the ego-lane based ROI method. Table 1 shows the comparison of two cases. It is obvious that ego-lane based ROI method has small average error compared with fixed ROI method. Table 1. Error mean and variance of each case Fixed ROI Lane ROI Error mean Error variance (a) Fixed ROI (b) Wrongly detected vehicle Fixed ROI Reference (a) Distance measurements using fixed ROI Ego-Lane based ROI Reference (b) Distance measurements using proposed ROI Fig. 11 Fixed ROI vs. ego-lane based ROI By investigating the curved road situation, it is analyzed how the proposed ego-lane based ROI improves the performance of preceding vehicle detection. Fig. 12(a) shows the example of fixed ROI and Fig 12(b) shows pixels having the same disparity as the output of preceding vehicle detection. Because the road is curved, the detected vehicle is located at adjacent road lane. Correct preceding vehicle, which should be used for longitudinal control, is rejected because of comparatively small peak height. In other sides, Fig. 12(c) shows the ROI established according to the recognized ego-lane. It is obviously shown that the ROI successfully captures the image of preceding vehicle ignoring the adjacent vehicle. (c) Adaptive ROI (d) Correctly detected vehicle Fig. 12 Comparison of ROI methods in case of road with curvature 6.3 Effect of Adaptive Threshold Preceding vehicle detection is tested with two thresholding methods: constant thresholding and adaptive thresholding. Fig. 13(a) shows the distance measurements calculated with constant thresholding. It can be observed that the distance measurements have very large error. Fig. 13(b) is the distance measurements calculated with adaptive thresholding. Table 2 shows that adaptive thresholding method has smaller average error than constant thresholding method. During experiments, constant threshold is fixed at a small value because only small threshold value can guarantee correct peak is included in candidate peaks. Therefore, a pixel belonging to background portion, which generally forms small peak at small disparity value, is detected as the preceding vehicle and causes false detection result. Error of constant threshold case includes background pixels and traffic markings. Table 2. Error mean and variance of each case Constant thresholding Adaptive thresholding Error mean Error variance Constant Threshold Reference (a) Distance measurements by constant thresholding

7 Adaptive Threshold Reference (c) Adaptive thresholding and detected peaks (b) Distance measurements by adaptive thresholding Fig. 13 Constance threshold vs. adaptive threshold By investigating a critical situation, it is analyzed how the proposed adaptive thresholding improves the performance of peak detection. Fig. 14(a) is the example of constance thresholding and detected peaks. A peak in circle denotes a peak recognized as preceding vehicle finally. In Fig. 14(b), it can be observed that traffic sign on the ground surface is pixels corresponding to the detected peak. Fig. 14(c) is the example of adaptive thresholding and detected peaks. Circled peak also denotes a peak recognized as preceding vehicle finally. In this case, thanks to the adaptive thresholding, correct peak, which is missed in the constance thresholding, can be successfully detected. Fig. 14(d) shows that the preceding vehicle causes pixels corresponding to the detected peak. (d) Correct detection result and corresponding pixels Fig. 14 Comparison of two thresholding methods 6.4 Comparison with Laser-Radar Comparing its output with the distance measured by laser radar validates the proposed method. It is also confirmed that large FOV (Field Of View) of vision system could improve the response time compared with the narrow FOV of laser radar Stereo Vision Laser Radar (a) Constant thresholding and detected peaks Fig. 15 Measured distances with laser radar and proposed method (b) Wrong detection result and corresponding pixels In an open space, test vehicle approaches a preceding vehicle standing still then returns to the initial position in backward direction. For the sake of safety, we fulfilled the experiment at our test track and could not use the ego-lane based ROI method. Fig. 15 shows the two distance sequences measured by proposed system and laser radar. Proposed system seems to

8 measure the distance correctly in the range of ~60m. Although the measured distance is not perfect, it looks sufficient for forward collision warning and avoidance. Furthermore, in the comparison between laser radar and proposed method, it should be noticed that vision based system can simultaneously detect ego-lane, which is necessary for lane keeping and target resolution on a curved road. Another major difference between stereo vision and laser radar is FOV. Laser radar should select narrow FOV to cover far distance because it uses TOF (Time Of Flight) principle without means of bearing angle measurement. In the contrast, vision system selects wider FOV because it can measure the bearing angle of object. Furthermore, Ego-lane based ROI eliminates the disturbance of vehicles on adjacent road lane. Fig.16 shows the distance measurements when a vehicle is going to cut-in. It is noticeable that vision system detects the cutting-in vehicle early compared with laser radar. Such a fast response time is crucial for the successful management of cut-in vehicle (a) Detected cutting-in vehicle Stereo Vision Laser Radar (b) Fast response of vision based system Fig. 16. Effect of vision system s wide FOV 7. CONCLUSION This paper proposes a stereo vision based obstacle detection, which consists of ego-lane based ROI establishment, feature based stereo matching, peak detection in disparity histogram by adaptive thresholding, edge/color based verification. Experiment results shows that the proposed method is valid in the range of ~60m. Effects of ego-lane based ROI and adaptive thresholding are analyzed. Comparison with laser radar shows that vision system has several advantages because of the capability of lane detection and wide FOV. REFERENCES [1] Bishop, R.: A Survey of Intelligent Vehicle Applications Worldwide, Proceedings Of IEEE Intelligent Vehicle Symposium, vol. 5, pp. 0-8, Dearborn, MI, USA (00). [2] Li, X., Liu, Z.Q., Leung, K.M.: Detection of vehicles from traffic scenes using fuzzy integrals, Pattern Recognition, vol. 35, no. 4, pp , (02). [3] Yu, M., Kim, Y.D.: An approach to Korean license plate recognition based on vertical edge matching, IEEE International Conference on Systems, Man, and Cybernetics, vol.4, pp , Nashville, TN, USA (00). [4] Ho Gi Jung, Dong Suk Kim, Pal Joo Yoon and Jai Hie Kim, Stereo Vision Based Localization of Free Parking Site, LNCS Vol (CAIP 05), pages , Sep. 05. (05) [5] Gern, A., Moebus, R., Franke, U.: Vision-based lane recognition under adverse weather conditions using optical flow, Intelligent Vehicle Symposium, vol. 2, pp , France (02). [6] Bertozzi, M., Broggi, A.: GOLD: A Parallel Real-Time Stereo Vision System for Generic Obstacle and Lane Detection, IEEE Transactions on Image Processing, vol. 7, no. 1, pp , (1998). [7] Dagan, E., Mano, O., Stein, G.P., Shashua, A.: Forward Collision Warning with a Single Camera, Intelligent Vehicles Symposium, pp , (04). [8] Paia, C.J., Tyanb, H.R., Liangc, Y.M., Liaoc, H.Y.M,: Pedestrian detection and tracking at crossroads, Pattern Recognition, vol. 37, pp , (04). [9] Escalera, A., Armingol, J.M., Mata, M.: Traffic sign recognition and analysis for intelligent vehicles, Image and Vision Computing, vol. 21, pp , (03). [] U. Rranke and I. Kutzbch, Fast Stereo based Object Detection for Stop&Go Traffic, Intelligent Vehicle, pp , [11] ml [12]

STEREO VISION-BASED FORWARD OBSTACLE DETECTION

STEREO VISION-BASED FORWARD OBSTACLE DETECTION International Journal of Automotive Technology, Vol. 8, No. 4, pp. 493 504 (2007) Copyright 2007 KSAE 1229 9138/2007/035 12 STEREO VISION-BASED FORWARD OBSTACLE DETECTION H. G. JUNG 1),2)*, Y. H. LEE 1),

More information

LIGHT STRIPE PROJECTION-BASED PEDESTRIAN DETECTION DURING AUTOMATIC PARKING OPERATION

LIGHT STRIPE PROJECTION-BASED PEDESTRIAN DETECTION DURING AUTOMATIC PARKING OPERATION F2008-08-099 LIGHT STRIPE PROJECTION-BASED PEDESTRIAN DETECTION DURING AUTOMATIC PARKING OPERATION 1 Jung, Ho Gi*, 1 Kim, Dong Suk, 1 Kang, Hyoung Jin, 2 Kim, Jaihie 1 MANDO Corporation, Republic of Korea,

More information

Structure Analysis Based Parking Slot Marking Recognition for Semi-automatic Parking System

Structure Analysis Based Parking Slot Marking Recognition for Semi-automatic Parking System Structure Analysis Based Parking Slot Marking Recognition for Semi-automatic Parking System Ho Gi Jung 1, 2, Dong Suk Kim 1, Pal Joo Yoon 1, and Jaihie Kim 2 1 MANDO Corporation Central R&D Center, Advanced

More information

Vision-based ACC with a Single Camera: Bounds on Range and Range Rate Accuracy

Vision-based ACC with a Single Camera: Bounds on Range and Range Rate Accuracy Vision-based ACC with a Single Camera: Bounds on Range and Range Rate Accuracy Gideon P. Stein Ofer Mano Amnon Shashua MobileEye Vision Technologies Ltd. MobileEye Vision Technologies Ltd. Hebrew University

More information

Semi-automatic Parking System Recognizing Parking Lot Markings

Semi-automatic Parking System Recognizing Parking Lot Markings Proceedings of AVEC 06 The 8 th International Symposium on Advanced Vehicle Control, August 20-24, 2006, Taipei, Taiwan AVEC060186 Semi-automatic Parking System Recognizing Parking Lot Markings Ho Gi Jung

More information

차세대지능형자동차를위한신호처리기술 정호기

차세대지능형자동차를위한신호처리기술 정호기 차세대지능형자동차를위한신호처리기술 008.08. 정호기 E-mail: hgjung@mando.com hgjung@yonsei.ac.kr 0 . 지능형자동차의미래 ) 단위 system functions 운전자상황인식 얼굴방향인식 시선방향인식 졸음운전인식 운전능력상실인식 차선인식, 전방장애물검출및분류 Lane Keeping System + Adaptive Cruise

More information

NOVEL USER INTERFACE FOR SEMI-AUTOMATIC PARKING ASSISTANCE SYSTEM

NOVEL USER INTERFACE FOR SEMI-AUTOMATIC PARKING ASSISTANCE SYSTEM F2006D130T NOVEL USER INTERFACE FOR SEMI-AUTOMATIC PARKING ASSISTANCE SYSTEM 1,2 Jung, Ho Gi, 1 Kim, Dong Suk *, 1 Yoon, Pal Joo, 2 Kim, Jaihie 1 MANDO Corporation, Republic of Korea, 2 Yonsei University,

More information

Stereo Image Rectification for Simple Panoramic Image Generation

Stereo Image Rectification for Simple Panoramic Image Generation Stereo Image Rectification for Simple Panoramic Image Generation Yun-Suk Kang and Yo-Sung Ho Gwangju Institute of Science and Technology (GIST) 261 Cheomdan-gwagiro, Buk-gu, Gwangju 500-712 Korea Email:{yunsuk,

More information

parco area delle Scienze, 181A via Ferrata, , Parma 27100, Pavia

parco area delle Scienze, 181A via Ferrata, , Parma 27100, Pavia Proceedings of the IEEE Intelligent Vehicles Symposium 2000 Dearbon (MI), USA October 3-5, 2000 Stereo Vision-based Vehicle Detection M. Bertozzi 1 A. Broggi 2 A. Fascioli 1 S. Nichele 2 1 Dipartimento

More information

Vehicle Detection Method using Haar-like Feature on Real Time System

Vehicle Detection Method using Haar-like Feature on Real Time System Vehicle Detection Method using Haar-like Feature on Real Time System Sungji Han, Youngjoon Han and Hernsoo Hahn Abstract This paper presents a robust vehicle detection approach using Haar-like feature.

More information

Integrated Vehicle and Lane Detection with Distance Estimation

Integrated Vehicle and Lane Detection with Distance Estimation Integrated Vehicle and Lane Detection with Distance Estimation Yu-Chun Chen, Te-Feng Su, Shang-Hong Lai Department of Computer Science, National Tsing Hua University,Taiwan 30013, R.O.C Abstract. In this

More information

Range Imaging Through Triangulation. Range Imaging Through Triangulation. Range Imaging Through Triangulation. Range Imaging Through Triangulation

Range Imaging Through Triangulation. Range Imaging Through Triangulation. Range Imaging Through Triangulation. Range Imaging Through Triangulation Obviously, this is a very slow process and not suitable for dynamic scenes. To speed things up, we can use a laser that projects a vertical line of light onto the scene. This laser rotates around its vertical

More information

LANE DEPARTURE WARNING SYSTEM FOR VEHICLE SAFETY

LANE DEPARTURE WARNING SYSTEM FOR VEHICLE SAFETY LANE DEPARTURE WARNING SYSTEM FOR VEHICLE SAFETY 1 K. Sravanthi, 2 Mrs. Ch. Padmashree 1 P.G. Scholar, 2 Assistant Professor AL Ameer College of Engineering ABSTRACT In Malaysia, the rate of fatality due

More information

Depth. Common Classification Tasks. Example: AlexNet. Another Example: Inception. Another Example: Inception. Depth

Depth. Common Classification Tasks. Example: AlexNet. Another Example: Inception. Another Example: Inception. Depth Common Classification Tasks Recognition of individual objects/faces Analyze object-specific features (e.g., key points) Train with images from different viewing angles Recognition of object classes Analyze

More information

Preceding vehicle detection and distance estimation. lane change, warning system.

Preceding vehicle detection and distance estimation. lane change, warning system. Preceding vehicle detection and distance estimation for lane change warning system U. Iqbal, M.S. Sarfraz Computer Vision Research Group (COMVis) Department of Electrical Engineering, COMSATS Institute

More information

Advanced Driver Assistance: Modular Image Sensor Concept

Advanced Driver Assistance: Modular Image Sensor Concept Vision Advanced Driver Assistance: Modular Image Sensor Concept Supplying value. Integrated Passive and Active Safety Systems Active Safety Passive Safety Scope Reduction of accident probability Get ready

More information

Pattern Feature Detection for Camera Calibration Using Circular Sample

Pattern Feature Detection for Camera Calibration Using Circular Sample Pattern Feature Detection for Camera Calibration Using Circular Sample Dong-Won Shin and Yo-Sung Ho (&) Gwangju Institute of Science and Technology (GIST), 13 Cheomdan-gwagiro, Buk-gu, Gwangju 500-71,

More information

2D rendering takes a photo of the 2D scene with a virtual camera that selects an axis aligned rectangle from the scene. The photograph is placed into

2D rendering takes a photo of the 2D scene with a virtual camera that selects an axis aligned rectangle from the scene. The photograph is placed into 2D rendering takes a photo of the 2D scene with a virtual camera that selects an axis aligned rectangle from the scene. The photograph is placed into the viewport of the current application window. A pixel

More information

STEREO-VISION SYSTEM PERFORMANCE ANALYSIS

STEREO-VISION SYSTEM PERFORMANCE ANALYSIS STEREO-VISION SYSTEM PERFORMANCE ANALYSIS M. Bertozzi, A. Broggi, G. Conte, and A. Fascioli Dipartimento di Ingegneria dell'informazione, Università di Parma Parco area delle Scienze, 181A I-43100, Parma,

More information

2 OVERVIEW OF RELATED WORK

2 OVERVIEW OF RELATED WORK Utsushi SAKAI Jun OGATA This paper presents a pedestrian detection system based on the fusion of sensors for LIDAR and convolutional neural network based image classification. By using LIDAR our method

More information

Research on the Measurement Method of the Detection Range of Vehicle Reversing Assisting System

Research on the Measurement Method of the Detection Range of Vehicle Reversing Assisting System Research on the Measurement Method of the Detection Range of Vehicle Reversing Assisting System Bowei Zou and Xiaochuan Cui Abstract This paper introduces the measurement method on detection range of reversing

More information

Measurement of Pedestrian Groups Using Subtraction Stereo

Measurement of Pedestrian Groups Using Subtraction Stereo Measurement of Pedestrian Groups Using Subtraction Stereo Kenji Terabayashi, Yuki Hashimoto, and Kazunori Umeda Chuo University / CREST, JST, 1-13-27 Kasuga, Bunkyo-ku, Tokyo 112-8551, Japan terabayashi@mech.chuo-u.ac.jp

More information

Lane Departure and Front Collision Warning Using a Single Camera

Lane Departure and Front Collision Warning Using a Single Camera Lane Departure and Front Collision Warning Using a Single Camera Huei-Yung Lin, Li-Qi Chen, Yu-Hsiang Lin Department of Electrical Engineering, National Chung Cheng University Chiayi 621, Taiwan hylin@ccu.edu.tw,

More information

EE795: Computer Vision and Intelligent Systems

EE795: Computer Vision and Intelligent Systems EE795: Computer Vision and Intelligent Systems Spring 2012 TTh 17:30-18:45 FDH 204 Lecture 14 130307 http://www.ee.unlv.edu/~b1morris/ecg795/ 2 Outline Review Stereo Dense Motion Estimation Translational

More information

Detecting and recognizing centerlines as parabolic sections of the steerable filter response

Detecting and recognizing centerlines as parabolic sections of the steerable filter response Detecting and recognizing centerlines as parabolic sections of the steerable filter response Petar Palašek, Petra Bosilj, Siniša Šegvić Faculty of Electrical Engineering and Computing Unska 3, 10000 Zagreb

More information

Fundamentals of Stereo Vision Michael Bleyer LVA Stereo Vision

Fundamentals of Stereo Vision Michael Bleyer LVA Stereo Vision Fundamentals of Stereo Vision Michael Bleyer LVA Stereo Vision What Happened Last Time? Human 3D perception (3D cinema) Computational stereo Intuitive explanation of what is meant by disparity Stereo matching

More information

Real-time Stereo Vision for Urban Traffic Scene Understanding

Real-time Stereo Vision for Urban Traffic Scene Understanding Proceedings of the IEEE Intelligent Vehicles Symposium 2000 Dearborn (MI), USA October 3-5, 2000 Real-time Stereo Vision for Urban Traffic Scene Understanding U. Franke, A. Joos DaimlerChrylser AG D-70546

More information

Transactions on Information and Communications Technologies vol 16, 1996 WIT Press, ISSN

Transactions on Information and Communications Technologies vol 16, 1996 WIT Press,   ISSN ransactions on Information and Communications echnologies vol 6, 996 WI Press, www.witpress.com, ISSN 743-357 Obstacle detection using stereo without correspondence L. X. Zhou & W. K. Gu Institute of Information

More information

On Road Vehicle Detection using Shadows

On Road Vehicle Detection using Shadows On Road Vehicle Detection using Shadows Gilad Buchman Grasp Lab, Department of Computer and Information Science School of Engineering University of Pennsylvania, Philadelphia, PA buchmag@seas.upenn.edu

More information

Mobile Human Detection Systems based on Sliding Windows Approach-A Review

Mobile Human Detection Systems based on Sliding Windows Approach-A Review Mobile Human Detection Systems based on Sliding Windows Approach-A Review Seminar: Mobile Human detection systems Njieutcheu Tassi cedrique Rovile Department of Computer Engineering University of Heidelberg

More information

Depth Estimation Using Monocular Camera

Depth Estimation Using Monocular Camera Depth Estimation Using Monocular Camera Apoorva Joglekar #, Devika Joshi #, Richa Khemani #, Smita Nair *, Shashikant Sahare # # Dept. of Electronics and Telecommunication, Cummins College of Engineering

More information

A New Algorithm for Shape Detection

A New Algorithm for Shape Detection IOSR Journal of Computer Engineering (IOSR-JCE) e-issn: 2278-0661,p-ISSN: 2278-8727, Volume 19, Issue 3, Ver. I (May.-June. 2017), PP 71-76 www.iosrjournals.org A New Algorithm for Shape Detection Hewa

More information

Change detection using joint intensity histogram

Change detection using joint intensity histogram Change detection using joint intensity histogram Yasuyo Kita National Institute of Advanced Industrial Science and Technology (AIST) Information Technology Research Institute AIST Tsukuba Central 2, 1-1-1

More information

A threshold decision of the object image by using the smart tag

A threshold decision of the object image by using the smart tag A threshold decision of the object image by using the smart tag Chang-Jun Im, Jin-Young Kim, Kwan Young Joung, Ho-Gil Lee Sensing & Perception Research Group Korea Institute of Industrial Technology (

More information

RECENTLY, customers have shown a growing interest in

RECENTLY, customers have shown a growing interest in 406 IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, VOL. 9, NO. 3, SEPTEMBER 2008 Scanning Laser Radar-Based Target Position Designation for Parking Aid System Ho Gi Jung, Member, IEEE, Young

More information

An Efficient Obstacle Awareness Application for Android Mobile Devices

An Efficient Obstacle Awareness Application for Android Mobile Devices An Efficient Obstacle Awareness Application for Android Mobile Devices Razvan Itu, Radu Danescu Computer Science Department Technical University of Cluj-Napoca Cluj-Napoca, Romania itu.razvan@gmail.com,

More information

Vision-based Frontal Vehicle Detection and Tracking

Vision-based Frontal Vehicle Detection and Tracking Vision-based Frontal and Tracking King Hann LIM, Kah Phooi SENG, Li-Minn ANG and Siew Wen CHIN School of Electrical and Electronic Engineering The University of Nottingham Malaysia campus, Jalan Broga,

More information

Sensor Fusion-Based Parking Assist System

Sensor Fusion-Based Parking Assist System Sensor Fusion-Based Parking Assist System 2014-01-0327 Jaeseob Choi, Eugene Chang, Daejoong Yoon, and Seongsook Ryu Hyundai & Kia Corp. Hogi Jung and Jaekyu Suhr Hanyang Univ. Published 04/01/2014 CITATION:

More information

Scene Text Detection Using Machine Learning Classifiers

Scene Text Detection Using Machine Learning Classifiers 601 Scene Text Detection Using Machine Learning Classifiers Nafla C.N. 1, Sneha K. 2, Divya K.P. 3 1 (Department of CSE, RCET, Akkikkvu, Thrissur) 2 (Department of CSE, RCET, Akkikkvu, Thrissur) 3 (Department

More information

Human Detection. A state-of-the-art survey. Mohammad Dorgham. University of Hamburg

Human Detection. A state-of-the-art survey. Mohammad Dorgham. University of Hamburg Human Detection A state-of-the-art survey Mohammad Dorgham University of Hamburg Presentation outline Motivation Applications Overview of approaches (categorized) Approaches details References Motivation

More information

Study on the Signboard Region Detection in Natural Image

Study on the Signboard Region Detection in Natural Image , pp.179-184 http://dx.doi.org/10.14257/astl.2016.140.34 Study on the Signboard Region Detection in Natural Image Daeyeong Lim 1, Youngbaik Kim 2, Incheol Park 1, Jihoon seung 1, Kilto Chong 1,* 1 1567

More information

Precision Roadway Feature Mapping Jay A. Farrell, University of California-Riverside James A. Arnold, Department of Transportation

Precision Roadway Feature Mapping Jay A. Farrell, University of California-Riverside James A. Arnold, Department of Transportation Precision Roadway Feature Mapping Jay A. Farrell, University of California-Riverside James A. Arnold, Department of Transportation February 26, 2013 ESRA Fed. GIS Outline: Big picture: Positioning and

More information

PMD [vision] Day Vol. 3 Munich, November 18, PMD Cameras for Automotive & Outdoor Applications. ifm electronic gmbh, V.Frey. Dr.

PMD [vision] Day Vol. 3 Munich, November 18, PMD Cameras for Automotive & Outdoor Applications. ifm electronic gmbh, V.Frey. Dr. R PMD [vision] Day Vol. 3 Munich, November 18, 2010 Dr. Volker Frey ifm electronic gmbh PMD Cameras for Automotive & Outdoor Applications Stand: 27.10.2010 Seite 1 I Working Principle PMD distance measurement

More information

ECE 172A: Introduction to Intelligent Systems: Machine Vision, Fall Midterm Examination

ECE 172A: Introduction to Intelligent Systems: Machine Vision, Fall Midterm Examination ECE 172A: Introduction to Intelligent Systems: Machine Vision, Fall 2008 October 29, 2008 Notes: Midterm Examination This is a closed book and closed notes examination. Please be precise and to the point.

More information

OCR For Handwritten Marathi Script

OCR For Handwritten Marathi Script International Journal of Scientific & Engineering Research Volume 3, Issue 8, August-2012 1 OCR For Handwritten Marathi Script Mrs.Vinaya. S. Tapkir 1, Mrs.Sushma.D.Shelke 2 1 Maharashtra Academy Of Engineering,

More information

ROBUST LINE-BASED CALIBRATION OF LENS DISTORTION FROM A SINGLE VIEW

ROBUST LINE-BASED CALIBRATION OF LENS DISTORTION FROM A SINGLE VIEW ROBUST LINE-BASED CALIBRATION OF LENS DISTORTION FROM A SINGLE VIEW Thorsten Thormählen, Hellward Broszio, Ingolf Wassermann thormae@tnt.uni-hannover.de University of Hannover, Information Technology Laboratory,

More information

Advanced Driver Assistance Systems: A Cost-Effective Implementation of the Forward Collision Warning Module

Advanced Driver Assistance Systems: A Cost-Effective Implementation of the Forward Collision Warning Module Advanced Driver Assistance Systems: A Cost-Effective Implementation of the Forward Collision Warning Module www.lnttechservices.com Table of Contents Abstract 03 Introduction 03 Solution Overview 03 Output

More information

Dense Image-based Motion Estimation Algorithms & Optical Flow

Dense Image-based Motion Estimation Algorithms & Optical Flow Dense mage-based Motion Estimation Algorithms & Optical Flow Video A video is a sequence of frames captured at different times The video data is a function of v time (t) v space (x,y) ntroduction to motion

More information

Stereo Vision. MAN-522 Computer Vision

Stereo Vision. MAN-522 Computer Vision Stereo Vision MAN-522 Computer Vision What is the goal of stereo vision? The recovery of the 3D structure of a scene using two or more images of the 3D scene, each acquired from a different viewpoint in

More information

Developing an intelligent sign inventory using image processing

Developing an intelligent sign inventory using image processing icccbe 2010 Nottingham University Press Proceedings of the International Conference on Computing in Civil and Building Engineering W Tizani (Editor) Developing an intelligent sign inventory using image

More information

Stereo imaging ideal geometry

Stereo imaging ideal geometry Stereo imaging ideal geometry (X,Y,Z) Z f (x L,y L ) f (x R,y R ) Optical axes are parallel Optical axes separated by baseline, b. Line connecting lens centers is perpendicular to the optical axis, and

More information

Character Recognition

Character Recognition Character Recognition 5.1 INTRODUCTION Recognition is one of the important steps in image processing. There are different methods such as Histogram method, Hough transformation, Neural computing approaches

More information

A Street Scene Surveillance System for Moving Object Detection, Tracking and Classification

A Street Scene Surveillance System for Moving Object Detection, Tracking and Classification A Street Scene Surveillance System for Moving Object Detection, Tracking and Classification Huei-Yung Lin * and Juang-Yu Wei Department of Electrical Engineering National Chung Cheng University Chia-Yi

More information

Real-Time Human Detection using Relational Depth Similarity Features

Real-Time Human Detection using Relational Depth Similarity Features Real-Time Human Detection using Relational Depth Similarity Features Sho Ikemura, Hironobu Fujiyoshi Dept. of Computer Science, Chubu University. Matsumoto 1200, Kasugai, Aichi, 487-8501 Japan. si@vision.cs.chubu.ac.jp,

More information

Image processing techniques for driver assistance. Razvan Itu June 2014, Technical University Cluj-Napoca

Image processing techniques for driver assistance. Razvan Itu June 2014, Technical University Cluj-Napoca Image processing techniques for driver assistance Razvan Itu June 2014, Technical University Cluj-Napoca Introduction Computer vision & image processing from wiki: any form of signal processing for which

More information

Machine learning based automatic extrinsic calibration of an onboard monocular camera for driving assistance applications on smart mobile devices

Machine learning based automatic extrinsic calibration of an onboard monocular camera for driving assistance applications on smart mobile devices Technical University of Cluj-Napoca Image Processing and Pattern Recognition Research Center www.cv.utcluj.ro Machine learning based automatic extrinsic calibration of an onboard monocular camera for driving

More information

Monitoring surrounding areas of truck-trailer combinations

Monitoring surrounding areas of truck-trailer combinations Monitoring surrounding areas of truck-trailer combinations Tobias Ehlgen 1 and Tomas Pajdla 2 1 Daimler-Chrysler Research and Technology, Ulm tobias.ehlgen@daimlerchrysler.com 2 Center of Machine Perception,

More information

ENY-C2005 Geoinformation in Environmental Modeling Lecture 4b: Laser scanning

ENY-C2005 Geoinformation in Environmental Modeling Lecture 4b: Laser scanning 1 ENY-C2005 Geoinformation in Environmental Modeling Lecture 4b: Laser scanning Petri Rönnholm Aalto University 2 Learning objectives To recognize applications of laser scanning To understand principles

More information

EE795: Computer Vision and Intelligent Systems

EE795: Computer Vision and Intelligent Systems EE795: Computer Vision and Intelligent Systems Spring 2012 TTh 17:30-18:45 WRI C225 Lecture 02 130124 http://www.ee.unlv.edu/~b1morris/ecg795/ 2 Outline Basics Image Formation Image Processing 3 Intelligent

More information

CONTRIBUTION TO THE INVESTIGATION OF STOPPING SIGHT DISTANCE IN THREE-DIMENSIONAL SPACE

CONTRIBUTION TO THE INVESTIGATION OF STOPPING SIGHT DISTANCE IN THREE-DIMENSIONAL SPACE National Technical University of Athens School of Civil Engineering Department of Transportation Planning and Engineering Doctoral Dissertation CONTRIBUTION TO THE INVESTIGATION OF STOPPING SIGHT DISTANCE

More information

Perceptual Quality Improvement of Stereoscopic Images

Perceptual Quality Improvement of Stereoscopic Images Perceptual Quality Improvement of Stereoscopic Images Jong In Gil and Manbae Kim Dept. of Computer and Communications Engineering Kangwon National University Chunchon, Republic of Korea, 200-701 E-mail:

More information

An Efficient Character Segmentation Based on VNP Algorithm

An Efficient Character Segmentation Based on VNP Algorithm Research Journal of Applied Sciences, Engineering and Technology 4(24): 5438-5442, 2012 ISSN: 2040-7467 Maxwell Scientific organization, 2012 Submitted: March 18, 2012 Accepted: April 14, 2012 Published:

More information

Last update: May 4, Vision. CMSC 421: Chapter 24. CMSC 421: Chapter 24 1

Last update: May 4, Vision. CMSC 421: Chapter 24. CMSC 421: Chapter 24 1 Last update: May 4, 200 Vision CMSC 42: Chapter 24 CMSC 42: Chapter 24 Outline Perception generally Image formation Early vision 2D D Object recognition CMSC 42: Chapter 24 2 Perception generally Stimulus

More information

Towards the completion of assignment 1

Towards the completion of assignment 1 Towards the completion of assignment 1 What to do for calibration What to do for point matching What to do for tracking What to do for GUI COMPSCI 773 Feature Point Detection Why study feature point detection?

More information

Prof. Fanny Ficuciello Robotics for Bioengineering Visual Servoing

Prof. Fanny Ficuciello Robotics for Bioengineering Visual Servoing Visual servoing vision allows a robotic system to obtain geometrical and qualitative information on the surrounding environment high level control motion planning (look-and-move visual grasping) low level

More information

Reference Tracking System for a Mobile Skid Steer Robot (Version 1.00)

Reference Tracking System for a Mobile Skid Steer Robot (Version 1.00) CENTER FOR MACHINE PERCEPTION CZECH TECHNICAL UNIVERSITY IN PRAGUE Reference Tracking System for a Mobile Skid Steer Robot (Version 1.00) Vladimír Kubelka, Vladimír Burian, Přemysl Kafka kubelka.vladimir@fel.cvut.cz

More information

Research Fellow, Korea Institute of Civil Engineering and Building Technology, Korea (*corresponding author) 2

Research Fellow, Korea Institute of Civil Engineering and Building Technology, Korea (*corresponding author) 2 Algorithm and Experiment for Vision-Based Recognition of Road Surface Conditions Using Polarization and Wavelet Transform 1 Seung-Ki Ryu *, 2 Taehyeong Kim, 3 Eunjoo Bae, 4 Seung-Rae Lee 1 Research Fellow,

More information

Miniaturized Camera Systems for Microfactories

Miniaturized Camera Systems for Microfactories Miniaturized Camera Systems for Microfactories Timo Prusi, Petri Rokka, and Reijo Tuokko Tampere University of Technology, Department of Production Engineering, Korkeakoulunkatu 6, 33720 Tampere, Finland

More information

Traffic Signs Recognition Experiments with Transform based Traffic Sign Recognition System

Traffic Signs Recognition Experiments with Transform based Traffic Sign Recognition System Sept. 8-10, 010, Kosice, Slovakia Traffic Signs Recognition Experiments with Transform based Traffic Sign Recognition System Martin FIFIK 1, Ján TURÁN 1, Ľuboš OVSENÍK 1 1 Department of Electronics and

More information

AUTOMATIC PARKING OF SELF-DRIVING CAR BASED ON LIDAR

AUTOMATIC PARKING OF SELF-DRIVING CAR BASED ON LIDAR AUTOMATIC PARKING OF SELF-DRIVING CAR BASED ON LIDAR Bijun Lee a, Yang Wei a, I. Yuan Guo a a State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University,

More information

arxiv: v1 [cs.cv] 28 Sep 2018

arxiv: v1 [cs.cv] 28 Sep 2018 Camera Pose Estimation from Sequence of Calibrated Images arxiv:1809.11066v1 [cs.cv] 28 Sep 2018 Jacek Komorowski 1 and Przemyslaw Rokita 2 1 Maria Curie-Sklodowska University, Institute of Computer Science,

More information

Compositing a bird's eye view mosaic

Compositing a bird's eye view mosaic Compositing a bird's eye view mosaic Robert Laganiere School of Information Technology and Engineering University of Ottawa Ottawa, Ont KN 6N Abstract This paper describes a method that allows the composition

More information

International Journal of Advance Engineering and Research Development

International Journal of Advance Engineering and Research Development Scientific Journal of Impact Factor (SJIF): 4.14 International Journal of Advance Engineering and Research Development Volume 3, Issue 3, March -2016 e-issn (O): 2348-4470 p-issn (P): 2348-6406 Research

More information

OBJECT detection in general has many applications

OBJECT detection in general has many applications 1 Implementing Rectangle Detection using Windowed Hough Transform Akhil Singh, Music Engineering, University of Miami Abstract This paper implements Jung and Schramm s method to use Hough Transform for

More information

A Longitudinal Control Algorithm for Smart Cruise Control with Virtual Parameters

A Longitudinal Control Algorithm for Smart Cruise Control with Virtual Parameters ISSN (e): 2250 3005 Volume, 06 Issue, 12 December 2016 International Journal of Computational Engineering Research (IJCER) A Longitudinal Control Algorithm for Smart Cruise Control with Virtual Parameters

More information

Visual Representation from Semiology of Graphics by J. Bertin

Visual Representation from Semiology of Graphics by J. Bertin Visual Representation from Semiology of Graphics by J. Bertin From a communication perspective Communication is too often taken for granted when it should be taken to pieces. (Fiske 91) Two basic schools

More information

Image Processing Fundamentals. Nicolas Vazquez Principal Software Engineer National Instruments

Image Processing Fundamentals. Nicolas Vazquez Principal Software Engineer National Instruments Image Processing Fundamentals Nicolas Vazquez Principal Software Engineer National Instruments Agenda Objectives and Motivations Enhancing Images Checking for Presence Locating Parts Measuring Features

More information

Stereo Vision Based Traversable Region Detection for Mobile Robots Using U-V-Disparity

Stereo Vision Based Traversable Region Detection for Mobile Robots Using U-V-Disparity Stereo Vision Based Traversable Region Detection for Mobile Robots Using U-V-Disparity ZHU Xiaozhou, LU Huimin, Member, IEEE, YANG Xingrui, LI Yubo, ZHANG Hui College of Mechatronics and Automation, National

More information

Forward Sensing System for LKS+ACC

Forward Sensing System for LKS+ACC SAE TECHNICAL PAPER SERIES 008-01-005 Forward Sensing System for LKS+ACC Ho Gi Jung, Yun Hee Lee and Pal Joo Yoon MANDO Corporation Jaihie Kim Yonsei University Reprinted From: Intelligent Vehicle Iniative

More information

Epipolar Geometry and Stereo Vision

Epipolar Geometry and Stereo Vision Epipolar Geometry and Stereo Vision Computer Vision Jia-Bin Huang, Virginia Tech Many slides from S. Seitz and D. Hoiem Last class: Image Stitching Two images with rotation/zoom but no translation. X x

More information

Integration of Multiple-baseline Color Stereo Vision with Focus and Defocus Analysis for 3D Shape Measurement

Integration of Multiple-baseline Color Stereo Vision with Focus and Defocus Analysis for 3D Shape Measurement Integration of Multiple-baseline Color Stereo Vision with Focus and Defocus Analysis for 3D Shape Measurement Ta Yuan and Murali Subbarao tyuan@sbee.sunysb.edu and murali@sbee.sunysb.edu Department of

More information

Self-calibration of an On-Board Stereo-vision System for Driver Assistance Systems

Self-calibration of an On-Board Stereo-vision System for Driver Assistance Systems 4-1 Self-calibration of an On-Board Stereo-vision System for Driver Assistance Systems Juan M. Collado, Cristina Hilario, Arturo de la Escalera and Jose M. Armingol Intelligent Systems Lab Department of

More information

Noise-resilient Road Surface and Free Space Estimation Using Dense Stereo

Noise-resilient Road Surface and Free Space Estimation Using Dense Stereo 2013 IEEE Intelligent Vehicles Symposium (IV) June 23-26, 2013, Gold Coast, Australia Noise-resilient Road Surface and Free Space Estimation Using Dense Stereo Jae Kyu Suhr, Member, IEEE, and Ho Gi Jung,

More information

CIRCULAR MOIRÉ PATTERNS IN 3D COMPUTER VISION APPLICATIONS

CIRCULAR MOIRÉ PATTERNS IN 3D COMPUTER VISION APPLICATIONS CIRCULAR MOIRÉ PATTERNS IN 3D COMPUTER VISION APPLICATIONS Setiawan Hadi Mathematics Department, Universitas Padjadjaran e-mail : shadi@unpad.ac.id Abstract Geometric patterns generated by superimposing

More information

Lane Markers Detection based on Consecutive Threshold Segmentation

Lane Markers Detection based on Consecutive Threshold Segmentation ISSN 1746-7659, England, UK Journal of Information and Computing Science Vol. 6, No. 3, 2011, pp. 207-212 Lane Markers Detection based on Consecutive Threshold Segmentation Huan Wang +, Mingwu Ren,Sulin

More information

Removing Shadows from Images

Removing Shadows from Images Removing Shadows from Images Zeinab Sadeghipour Kermani School of Computing Science Simon Fraser University Burnaby, BC, V5A 1S6 Mark S. Drew School of Computing Science Simon Fraser University Burnaby,

More information

CLASSIFICATION FOR ROADSIDE OBJECTS BASED ON SIMULATED LASER SCANNING

CLASSIFICATION FOR ROADSIDE OBJECTS BASED ON SIMULATED LASER SCANNING CLASSIFICATION FOR ROADSIDE OBJECTS BASED ON SIMULATED LASER SCANNING Kenta Fukano 1, and Hiroshi Masuda 2 1) Graduate student, Department of Intelligence Mechanical Engineering, The University of Electro-Communications,

More information

DISTANCE MEASUREMENT USING STEREO VISION

DISTANCE MEASUREMENT USING STEREO VISION DISTANCE MEASUREMENT USING STEREO VISION Sheetal Nagar 1, Jitendra Verma 2 1 Department of Electronics and Communication Engineering, IIMT, Greater Noida (India) 2 Department of computer science Engineering,

More information

Stochastic Road Shape Estimation, B. Southall & C. Taylor. Review by: Christopher Rasmussen

Stochastic Road Shape Estimation, B. Southall & C. Taylor. Review by: Christopher Rasmussen Stochastic Road Shape Estimation, B. Southall & C. Taylor Review by: Christopher Rasmussen September 26, 2002 Announcements Readings for next Tuesday: Chapter 14-14.4, 22-22.5 in Forsyth & Ponce Main Contributions

More information

EE368 Project Report CD Cover Recognition Using Modified SIFT Algorithm

EE368 Project Report CD Cover Recognition Using Modified SIFT Algorithm EE368 Project Report CD Cover Recognition Using Modified SIFT Algorithm Group 1: Mina A. Makar Stanford University mamakar@stanford.edu Abstract In this report, we investigate the application of the Scale-Invariant

More information

Segmentation and Tracking of Partial Planar Templates

Segmentation and Tracking of Partial Planar Templates Segmentation and Tracking of Partial Planar Templates Abdelsalam Masoud William Hoff Colorado School of Mines Colorado School of Mines Golden, CO 800 Golden, CO 800 amasoud@mines.edu whoff@mines.edu Abstract

More information

Sensory Augmentation for Increased Awareness of Driving Environment

Sensory Augmentation for Increased Awareness of Driving Environment Sensory Augmentation for Increased Awareness of Driving Environment Pranay Agrawal John M. Dolan Dec. 12, 2014 Technologies for Safe and Efficient Transportation (T-SET) UTC The Robotics Institute Carnegie

More information

Extended Floating Car Data System - Experimental Study -

Extended Floating Car Data System - Experimental Study - 2011 IEEE Intelligent Vehicles Symposium (IV) Baden-Baden, Germany, June 5-9, 2011 Extended Floating Car Data System - Experimental Study - R. Quintero, A. Llamazares, D. F. Llorca, M. A. Sotelo, L. E.

More information

CITS 4402 Computer Vision

CITS 4402 Computer Vision CITS 4402 Computer Vision A/Prof Ajmal Mian Adj/A/Prof Mehdi Ravanbakhsh, CEO at Mapizy (www.mapizy.com) and InFarm (www.infarm.io) Lecture 02 Binary Image Analysis Objectives Revision of image formation

More information

Lecture 19: Depth Cameras. Visual Computing Systems CMU , Fall 2013

Lecture 19: Depth Cameras. Visual Computing Systems CMU , Fall 2013 Lecture 19: Depth Cameras Visual Computing Systems Continuing theme: computational photography Cameras capture light, then extensive processing produces the desired image Today: - Capturing scene depth

More information

Real-Time Detection of Road Markings for Driving Assistance Applications

Real-Time Detection of Road Markings for Driving Assistance Applications Real-Time Detection of Road Markings for Driving Assistance Applications Ioana Maria Chira, Ancuta Chibulcutean Students, Faculty of Automation and Computer Science Technical University of Cluj-Napoca

More information

MATHEMATICAL IMAGE PROCESSING FOR AUTOMATIC NUMBER PLATE RECOGNITION SYSTEM

MATHEMATICAL IMAGE PROCESSING FOR AUTOMATIC NUMBER PLATE RECOGNITION SYSTEM J. KSIAM Vol.14, No.1, 57 66, 2010 MATHEMATICAL IMAGE PROCESSING FOR AUTOMATIC NUMBER PLATE RECOGNITION SYSTEM SUNHEE KIM, SEUNGMI OH, AND MYUNGJOO KANG DEPARTMENT OF MATHEMATICAL SCIENCES, SEOUL NATIONAL

More information

Real Time Stereo Vision Based Pedestrian Detection Using Full Body Contours

Real Time Stereo Vision Based Pedestrian Detection Using Full Body Contours Real Time Stereo Vision Based Pedestrian Detection Using Full Body Contours Ion Giosan, Sergiu Nedevschi, Silviu Bota Technical University of Cluj-Napoca {Ion.Giosan, Sergiu.Nedevschi, Silviu.Bota}@cs.utcluj.ro

More information

Application of Geometry Rectification to Deformed Characters Recognition Liqun Wang1, a * and Honghui Fan2

Application of Geometry Rectification to Deformed Characters Recognition Liqun Wang1, a * and Honghui Fan2 6th International Conference on Electronic, Mechanical, Information and Management (EMIM 2016) Application of Geometry Rectification to Deformed Characters Liqun Wang1, a * and Honghui Fan2 1 School of

More information

3D-OBJECT DETECTION METHOD BASED ON THE STEREO IMAGE TRANSFORMATION TO THE COMMON OBSERVATION POINT

3D-OBJECT DETECTION METHOD BASED ON THE STEREO IMAGE TRANSFORMATION TO THE COMMON OBSERVATION POINT 3D-OBJECT DETECTION METHOD BASED ON THE STEREO IMAGE TRANSFORMATION TO THE COMMON OBSERVATION POINT V. M. Lisitsyn *, S. V. Tikhonova ** State Research Institute of Aviation Systems, Moscow, Russia * lvm@gosniias.msk.ru

More information