Non-iterative Construction of Super-Resolution Image from an Acoustic Camera Video Sequence

Size: px
Start display at page:

Download "Non-iterative Construction of Super-Resolution Image from an Acoustic Camera Video Sequence"

Transcription

1 CIHSPS IEEE Internatonal Conference on Computatonal Intellgence for Homeland Securty and Personal Safety Orlando, FL, USA, 3 March Aprl 005 Non-teratve Constructon of Super-Resoluton Image from an Acoustc Camera Vdeo Sequence K. Km, N. Nerett, N. Intrator Insttute for Bran and Neural Systems, Brown Unversty, Provdence RI 09, USA Phone: +40) , Fax: +40) , E mal: ko@brown.edu. Abstract The low vsblty condton of underwater envronments s often a lmtng factor n the exploraton of rvers or coastal areas. In such stuatons, relatvely hgh resoluton mages produced by an acoustc camera can be an nformatve measure for the detecton of terran and objects. In ths paper, we present a non-teratve super-resoluton algorthm for the constructon of a hgh resoluton mage from multple frames of acoustc camera mages. We dscuss the geometry of nsonfcaton and mage acquston of an acoustc camera, and descrbe the detaled procedures for the regstraton, nsonfcaton correcton, and fuson. Keywords acoustc camera, mosacng, regstraton, vdeo enhancement, super-resoluton. I. INTRODUCTION An acoustc camera s a hgh resoluton ultrasonc magng devce that can produce underwater vdeo sequences, very useful for the purpose of the survellance of coastal areas []. It can be mounted on an autonomous underwater vehcle AUV) or carred by a dver, and can be sent out to explore potentally harmful objects. Despte the merts of acoustc cameras over other sonar systems, t stll has shortcomngs compared to normal optcal cameras: ) Lmtaton of sght range: Unlke optcal cameras whch have a -D array of photosensors, acoustc cameras have a - D transducer array. -D representaton s obtaned from the temporal sequence of the transducer array. For ths reason, t can collect nformaton from a lmted range. ) Low sgnal-to-nose rato SNR): The transducer sze s comparable to the wavelength of ultrasonc waves, so the ntensty of a pxel depends not only on the ampltude, but also on the phase dfference of the reflected sgnal. For ths reason, the presence of background nose n underwater envronments results n a Rcan dstrbuton of nose observed n ultrasound mages []. The SNR s also sgnfcantly lower than n optcal mages because of the transducer sze. ) Low resoluton wth respect to optcal mages: Due to the large scale of the wavelength of ultrasound compared to the wavelength of lght, the number of pxels n the horzontal axs s lmted. v) Inhomogeneous nsonfcaton: Snce one dmenson of the mage s acqured merely based on the relatve sgnal arrval tmes, a sngle nsonfer was used. Consequently, due to the ansotropy of ultrasound radaton from the nsonfer, the nsonfcaton s nhomogeneous n acoustc camera mages. The above lmtatons can be addressed by mage mosacng, whch s broadly used to buld a wder vew mage [3], [4], [5], or to estmate the moton of a vehcle [6], [7]. For ordnary mages, mosacng s also used for mage enhancement such as denosng, deblurrng, or super-resoluton [8], [9]. In ths paper, we descrbe the mosacng of a sequence of acoustc camera mages, and present a new non-teratve algorthm for the constructon of a hgh resoluton mage from the mage sequence. II. IMAGING GEOMETRY The transformaton between two acoustc camera mages can be calculated by puttng one mage nto the coordnate system where the mage s on the xy-plane wth the postve y-axs along the center lne of the mage and the center of the arc at the orgn Fg. ). Durng the magng process, a pont denoted by a poston vector x = x, y, z) s projected to the polar coordnates r, a) as follows: r = x, α = sn x r xy, ) where r xy x + y ) /, or to the Cartesan coordnates u = u, v),

2 u and v can then be rewrtten as z = ax + by + z 0. 5) u z ) = + x + y { R + R 3 a) x + R + R 3 b) y Fg.. The magng geometry of an acoustc camera. The camera s located at the orgn of the xyz-coordnate system wth the ptch, yaw, and roll 0. In the next frame x y z -coordnate), the camera s dsplaced by δx = δx, δy, δz) and rotated by φ, θ, ψ). u = r sn α = x + tan β ) v = r cos α = y + tan β, 3) where β s the angle between x and the magng plane. When the camera s translated by δx = δx, δy, δz) and rotated by φ, θ, ψ), the new coordnates of x are x = x, y, z ) = R φθψ x δx) 4) where R φθψ s a 3 3 rotaton matrx. The lnear transformaton T between two mages should satsfy u v = = T = T x + tan β y + tan β x + tan β y + tan β u v, where tan β = z /x + y ) /. When the reflectng ponts of the target object are located roughly on a plane such as the sea floor, z can be approxmated by R φθψ = cos φ cos ψ cos φ sn ψ sn φ cos θ sn φ sn θ sn ψ sn φ sn θ cos ψ sn φ cos ψ sn φ sn ψ cos φ cos θ + cos φ sn θ sn ψ cos φ sn θ cos ψ cos θ sn ψ sn θ cos θ cos ψ.. R δx + R δy + R 3 δz z 0 )) }, v z ) = + x + y { R + R 3 a) x + R + R 3 b) y R δx + R δy + R 3 δz z 0 )) }, where a, b, and z /x + y ) / are suffcently small that ther squares are neglgble. For example, the maxmum devaton angle β max of a DIDSON system s 5, thus + tan β max ) / = Up to a frst order of approxmaton, we have, T = R R + R 3 a R + R 3 b δx + R δy +R 3 δz z 0 )) R R + R 3 a R + R 3 b δx + R δy +R 3 δz z 0 )) 0 0.6) Ths serves as a frst order approxmaton of the transform between two acoustc camera mages. Further approxmaton wll be studed n subsequent work by segmentng the mage nto local planes dependng on the levels of elevaton. The sx unknown parameters of the affne transform can be obtaned by matchng features n two mages. However, other parameters such as R j, a, b, or δx n 6) cannot be fgured out separately because those parameters are coupled and underconstraned. Consequently, under the above approxmaton, t s mpossble to reconstruct the precse egomoton of the acoustc camera merely based on mage regstraton parameters. III. METHODOLOGY The typcal four steps of mage regstraton are: feature detecton, feature matchng, transformaton estmaton, and mage resamplng and transformaton [0]. Feature detecton s the process of fndng objects such as corners, edges, lne ntersectons, etc. from mages, manually or automatcally. The features from the sensed mage are pared wth the correspondng features n the reference mage n the second step. In the thrd step, the transformaton s estmated based on the dsplacement vector of each feature. Once the mappng s establshed, the multple mages are combned to generate a fuson mage.

3 In our work, we have found that hgh curvature ponts can be useful as features of nterest n acoustc camera mages. The sum of squared dfference s used to measure the dssmlarty between two mages n the second step. Transform parameters are estmated va a random samplng based method. After the parameters of the affne transform are obtaned, all mages are combned by weghted average. A. Coordnate mappng and nhomogeneous nsonfcaton equalzaton In order to restore the spatal homogenety of the mage, a transformaton to the Cartesan coordnates has to be performed. Due to the fact that the feld of vew n the angular coordnate of dfferent sensors does not overlap, the resultng pxel sze n the Cartesan coordnates s not homogeneous. Therefore, nearest neghbor nterpolaton was appled to fll the gaps n the mage n the Cartesan coordnate system. Due to the acoustc acquston of mages whch was performed by nsonfyng the area wth a sngle source, an nhomogeneous ntensty profle s obtaned. Ths has to be corrected for effcent mosacng. Prevous work on separaton of llumnaton from reflectance has been based on Retnex theory [], []. Usng a homomorphc flterng method [3], one estmates the nsonfcaton of an mage, and reconstructs the mage under unform nsonfcaton. Modelng the nsonfcaton as a homomorphc flter, the mage produced by the sonar system can be wrtten as Ĩ = L M Î + η 7) where Ĩ s the -th observed mage, L the -th nsonfcaton operaton, M the -th down-samplng operaton, Î the ground truth mage under unform nsonfcaton, and η a Gaussan nose. The estmated nsonfcaton ntensty L s calculated by applyng ) a Gaussan flter to the orgnal mage, L = dag HĨ, where H s a Gaussan blurrng operaton. The estmated unform nsonfcaton mage s Ĩ u) = L L M Î + η) M Î + L η. 8) The sum of squared dfference between the -th and j-th unform nsonfcaton mages s SSD,j = Ĩ u) Ĩ u) j + L η L η 9) and can be used as a ds-smlarty measure. The second term n 9) s ndependent of the true mage, and may be regarded as a constant, provded the nose s unform. A regularzaton factor may be added to L to prevent erroneously excessve ntensty from the speckles n low nsonfcaton regons. B. Feature detecton and putatve matchng Feature detecton and matchng are computatonally demandng. A Gaussan pyramd has been proposed as a multscale approach for effcent feature detecton and matchng [4], [0]. A pxel n the polar coordnates mage corresponds to or several pxels n the mapped mage. For example, n an mage wth the range m, the number of pxels that correspond to a sngle pxel n the polar coordnates mage vared from to 8. Magnfed pxels result n jagged edges n the mapped mage. In our mages, feature detecton at the thrd level of the Gaussan pyramd reduces false detecton of corners at the jagged edges. Feature detecton and putatve matchng s ntalzed by translatonal dsplacement detecton. Translatonal dsplacement between the sensed mage and the reference mage s calculated by an exhaustve search on the fourth level of the Gaussan pyramd. Ths process drastcally reduces the area of exhaustve search. After translaton s estmated, hgh curvature ponts of the sensed mage are detected usng the Harrs corner detector [5]. The second moment matrx M s computed usng the followng relatonshp: M = e x x/σ s I) I) ), 0) where σ s s the scale factor of corners, and I s the gradent vector of the mage. The response after the Harrs corner detecton s, R = det M ktrm), ) where k s set to The local maxma of R correspond to corners. The corners are matched to the correspondng ponts n the reference mage by another exhaustve search on the thrd level of the Gaussan pyramd. C. Transform estmaton Image changes due to the sonar system movement are modeled by an affne transformaton as derved n the secton II. The affne transformaton descrbes the mage changes by yaw, small ptch and roll and translatonal movement of the sonar system. Ths s vald when multple objects are not present at the same range and angle, whch s the case wth the great majorty of mages n our dataset []. The detaled procedure of the algorthm s as follows: ) Feature ponts estmaton: Usng the Harrs corner detector, compute 50 nterest ponts n a preprocessed acoustc camera mage. ) Correspondng ponts search: For a square patch around each feature pont n the sensed mage, fnd the sub-pxel-wse dsplacement n the next mage, usng a cross-correlaton based matchng. ) Transform parameter estmaton: Repeat the followng )-3) for 000 samples.

4 ) Select 3 putatve matchng pars. ) Usng the matchng pars, estmate the parameters of the affne transform. 3) Fnd the nlers of the estmated transform, and repeat ) wth the nlers untl the estmated nlers are stablzed. v) Set a certan k percentle to defne a threshold n of feature ponts. Then, fnd the n pars of ponts that are closest to each other. The least mean squared error of the pars s used as the crteron. We use the crteron of least square error of k% of samples, where k s emprcally determned. It s smlar to the leastmedan of squares LMS) [6], but t dffers n that t can have a lower breakdown pont k nstead of 0.5 of LMS), and t uses the mean squared error nstead of the k-th percentle as the measure of error. It works well wth only a few feature pont pars wth hgh percentage of outlers. In addton, t yelds a measure of goodness of the transformaton, whch helps to decde whether to contnue mosacng or to stop, for example when the rsk of msmatch s hgh. D. Mosacng After regstraton, a mosac mage s constructed. Snce the nose s present regardless of the nsonfcaton condton, t can deterorate the mosac mage f not treated properly. For example, f we average well-nsonfed mages and poorly-nsonfed mages, the SNR wll become lower because nose wll be accumulated. A unformly nsonfed mage s, when addtve nose η of a standard devaton σ s added, Î + η σ ) and the SNR s smply /σ, and when t s nsonfed wth nhomogeneous nsonfcaton L, L Î + η σ 3) the SNR s σ dagl ). Mosacng va unform averagng can be descrbed as the followng relatonshp: Ĩ u) mosac = N the SNR of the mosac mage s N Ĩ, 4) = SNR u) mosac = σ N L. 5) Note that the SNR s a column vector of the same sze as ˆL because the nsonfcaton ntensty vares wthn the mage. A desrable stuaton s when poorly nsonfed regons receve lower weght n the averagng. Denote the weght of the -th mage by α, where α = I, and I s the dentty matrx. Then, the mosac mage produced va nhomogeneous weghtng s, of whch the SNR, SNR w) mosac s, Ĩ w) mosac = α Ĩ 6) SNR w) mosac α,..., α N ) = ) / α σ α L 7) ) / L σ. 8) Equalty holds n 8) when α k = L ) L k. Thus, the maxmum SNR of the mosac mage s acheved when the transformed mages are combned as follows: Ĩ w) mosac = L ) L I. 9) E. Maxmum-Lkelhood estmaton of equalzed mage Together wth the weghted mosac mage, the mage qualty can be enhanced by an addtonal process. Accordng to the nhomogeneous nsonfcaton model, the ntensty of a reflected sgnal vares dependng on the nsonfcaton condton. Gven the observed pxel values Ĩ, Ĩ,...Ĩ N under the nsonfcaton ntensty L, L,...L N, the estmated true reflectvty mage s Ĩ that maxmzes P Ĩ Ĩ, Ĩ,...Ĩ N ) = P Ĩ, Ĩ,...Ĩ N Ĩ)P Ĩ) P Ĩ, Ĩ,...Ĩ N ) 0) and, by approxmatng the Rcan nose dstrbuton by a Gaussan dstrbuton [7], we have the maxmum lkelhood mage Ĩ mle = arg max ln P Ĩ, Ĩ,...Ĩ N Ĩ) Ĩ = arg max { Ĩ σ L L Ĩ Ĩ) +... ) } +L N L N ĨN Ĩ where the denomnator of 0) s neglected. By combnng the exponents, fnally, where Ĩ mle = arg max ) Ĩ0 Ĩ σ0 Ĩ ) σ 0 = σ L, ) Ĩ 0 = L ) L Ĩ. 3)

5 Ths result s useful snce we can choose an approprate pror P Ĩ) to get the best a posteror estmaton of the equalzed mage. Note that the maxmum-lkelhood estmaton result has the same weght as n 9), and the only dfference s that the sum of the nsonfcaton ntensty n 9) s replaced by the square sum n 3). Therefore, ths maxmum lkelhood estmaton mage has the same maxmum SNR property as the weghted mosac mage n 9). F. Estmaton of the Pror from Pont Spread Functon A more detaled procedure of mage formaton from an magng devce s typcally modelled by the followng equaton: Ĩ = L M P Î + η where Î represents the true or hgh-resoluton mage, P the -th pont spread functon operaton, M the down-samplng operaton, L the llumnaton or nsonfcaton for sonar systems), η the addtve nose, and Ĩ the -th observed mage. The process pont spread functon PSF) and downsamplng operaton fuse multple pxels n the hgh resoluton mage nto one pxel n the low resoluton mage. Usually, the fuson s performed wth dfferent spatal weghts gven to the hgh resoluton pxels, whch s n most of the cases modeled by Gaussan dstrbutons. The fuson process can be represented as the followng: θ w = θ 4)... ) w =, 5) where θ s a low resoluton pxel value, w the weght vector, θ a vector of the pxels n the hgh resoluton mage that correspond to the pxel n the low resoluton mage. On the other hand, we have another constrant to estmate the probablty densty functon θ, whch s that a pxel value s not ndependent of ts adjacent pxel values. A Gaussan Markov Random Feld s one of the popular models for ths property, whch gves us the followng probablty densty functon of θ: ) P θ) = N exp θ D D)θ, 6) where D s an operaton of frst dervatve, and N a normalzaton constant. Combnng those two constrants, we obtan the pror, and the fnal maxmum a posteror MAP) mage can be obtaned: ) Ĩ map = L P L P Ĩ, 7) where P s the pont spread functon operated onto the ground truth mage. IV. RESULTS The algorthm was tested on a boat wreckage sequence. A DIDSON system [] scanned a shpwreck at approxmately 30 meters depth for 85 seconds and took 446 frames of mages. About 40 frames among them show the vessel from head to stern, and another 40 frames show t from stern to head. We appled the algorthm to those two sub-sequences, and bult two mosac mages. Fg. depcts two consecutve acoustc mages together wth a set of matched crcle) and non-matched trangle) feature ponts. These matched feature ponts n the reference mage, whch were found usng the cross-correlaton of patches around the feature ponts n the sensed mage, are used to estmate the geometrc transformaton between the two mages. Fg.. Matched crcle) and non-matched trangle) feature ponts whch were obtaned from the thrd level of the Gaussan pyramd usng the Harrs corner detector. Features from a sensed mage are pared wth correspondng ponts n the reference mage. A 5 5 patch around each feature pont n the sensed mage s matched wth the same szed patch from the correspondng area n the reference mage. The outlers features wth weaker matchng) are defned by those pars wth hgher matchng error after the estmated transformaton. Cross-correlaton was found to be more robust than a more conventonal approach n whch features are ndependently found and matched between the two mages. Ths s a consequence of the hgh nose n the mage and the fact that the exact locaton of the feature s not so well defned. Fg. 3 represents the man result of the paper,.e., a mosac mage of multple acoustc mages. The mosaced mage contans nformaton whch spans multple frames, each frame correspondng to a small porton of the nsonfed object. The combnaton mages whch have been transformed to be n the same coordnate system provde subpxel mage resoluton enhancement. Fg. 4a) shows a frame of another shpwreck target before mosacng. The resoluton enhancement follows from the fact that one pxel n the orgnal polar coordnate system s mapped to multple pxels wth the same ntensty n the Cartesan coordnate system. Dfferent frames lead to partal overlap of these The data has been provded by E. O. Belcher from Appled Physcs Laboratory, Unversty of Washngton under ONR support.

6 multple pxels, so that after averagng a subpxel resoluton s acheved See Fg. 4a)). Averagng of dfferent acoustc mages after brngng them to the same coordnate system same vewpont) leads to the classcal effect of denosng. Ths s clearly seen n Fg. 3 on the whole target, and n partcular n the comparson of a detaled structure of the other target n Fg. 4. The top two panels of Fg. 3 depct a mosaced mage from the same sequence of acoustc mages. In panel b), the nsonfcaton profle was utlzed durng averagng. Panel c) and d) represent the same target from a dfferent acoustc mage sequence wth panel d) utlzng the nsonfcaton profle durng averagng. a) b) V. CONCLUSION Acoustc camera technology s becomng essental for underwater exploraton, and has the potental for the applcaton to other areas for vsual nvestgatons. The acoustc camera, wth ts specfc sensor desgn, poses some challenges n terms of mage resoluton, nose removal and area coverage. In ths paper, we have presented a complete algorthm to acheve mage mosacng, denosng and resoluton enhancement from a sequence of acoustc camera mages. We descrbed the steps that were requred to acheve ths mosacng. Ths ncluded modelng the specfc geometry of acoustc camera mages whch sharply dffers from pnhole camera geometry. The dfferent geometry, and n partcular, the fact that the mages are acqured n a polar coordnate system, complcates the search and matchng of feature ponts n consecutve mages. Moreover, n ths partcular geometry, pxels n the polar coordnate system are mapped to a collecton of pxels wth the same ntensty n the Cartesan coordnate system. Snce consecutve mages were taken from dfferent vewponts, a subpxel enhancement effect was acheved n the process of averagng n addton to the denosng effect. We have presented a novel method n whch features extracted by a certan algorthm are locally matched to the reference mage va cross-correlaton. Ths method was found to be more robust than a conventonal approach n whch features are ndependently found and matched between two mages. In partcular, ths s more pronounced when the number of pxels avalable for feature comparson s lmted. The dfferent geometry, and n partcular the fact that the mages are acqured n a polar coordnate system complcates the search and matchng of feature ponts n consecutve mages. Moreover, n ths partcular geometry, pxels n the polar coordnate system are mapped to a collecton of pxels wth the same ntensty n the Cartesan coordnate system. Snce consecutve mages were taken from dfferent vewponts, a subpxel enhancement effect was acheved n the process of c) d) Fg. 3. Demonstraton of weghted averagng effect: Mosacng was performed wth 38 mages whch were averaged after the correspondng moton compensaton transformaton was appled to each of them. Panel a) demonstrates the unform average of the whole mages, whle panel b) demonstrates the weghted average, n whch nsonfcaton profle was utlzed durng averagng. See secton III. D for detals.) Panel c) and d) depct the same target from a dfferent mage sequence. averagng n addton to the denosng effect. We have presented a method n whch features are found n a one mage, and are then locally matched to the another mage va crosscorrelaton. ACKNOWLEDGMENT Ths work was partly supported by ONR grant N C-096. We thank E. O. Belcher for answerng all our questons about the data. We would lke to send Professor Leon N Cooper and other members of IBNS for provdng valuable comments.

7 a) Fg. 4. Resoluton enhancement by weghted averagng mages. a) A sngle frame mage. b) A mosac mage of 7 consecutve frames followed by a geometrc transformaton. See Secton III. F.) b) REFERENCES [] E. O. Belcher, B. Matsuyama, and G. M. Trmble, Object dentfcaton wth acoustc lenses, n Proceedngs of Oceans 0 MTS/IEEE, 00, pp. 6. [] R. F. Wagner, S. W. Smth, J. M. Sandrk, and H. Lopez, Statstcs of speckle n ultrasound b-scans, IEEE Transactons on Soncs and Ultrasoncs, vol. 30, no. 3, pp , 983. [3] R. L. Marks, S. M. Rock, and M. J. Lee, Real-tme vdeo mosackng of the ocean floor, IEEE Journal of Oceanc Engneerng, vol. 0, no. 3, pp. 9 4, 995. [4] Y. Rzhanov, L. M. Lnnet, and R. Forbes, Underwater vdeo mosacng for seabed mappng, n Internatonal Conference on Image processng, 000, vol., pp [5] U. Castellan, A. Fusello, and V. Murno, Regstraton of multple acoustc range vews for underwater scene reconstructon, Computer Vson and Image Understandng, vol. 87, pp , 00. [6] R. García, X. Cufí, and L. Pacheco, Image mosackng for estmatng the moton of an underwater vehcle, n 5th IFAC Conference on Manoeuvreng and Control of Marne Craft, 000. [7] S. Negahdarpour, X. Xu, and A. Khamene, A vson system for realtme postonng, navgaton and vdeo mosacng of sea floor magery n the applcaton of rovs/auvs, n IEEE Workshop on Applcatons of Computer Vson, 998, pp [8] D. Capel and A. Zsserman, Computer vson appled to superresoluton, IEEE Sgnal Processng Magazne, vol. 0, no. 3, pp. 7 86, 003. [9] A. Smolc and T. Wegand, Hgh-resoluton vdeo mosacng, n Internatonal Conference on Image Processng, 00, vol. 3, pp [0] B. Ztová and J. Flusser, Image regstraton methods: a survey, Image and Vson Computng, vol., pp , 003. [] E. H. Land, Recent advances n the retnex theory and some mplcatons for cortcal computatons: Color vson and the natural mage, PNAS, vol. 80, pp , 983. [] E. H. Land and J. J. McCann, Lghtness and the retnex theory, Journal of the Optcal Socety of Amerca, vol. 6, pp., 97. [3] E. H. Land, An alternatve technque for the computaton of the desgnator n the retnex theory of color vson, PNAS, vol. 83, pp , 986. [4] D. A. Forsyth and J. Ponce, Computer Vson: A Modern Approach, Pearson Educaton, Inc., Upper Saddle Rver, NJ, 003. [5] C. J. Harrs and M. Stephens, A combned corner and edge detector, n Proceedngs on 4th Alvey Vson Conference, 988, pp [6] P. J. Rousseeuw, Least medan of square regresson, Journal of the Amercan Statstcal Assocaton, vol. 79, pp , 984. [7] J. Sjbers, J. den Dekker, P. Scheunders, and Drk Van Dyck, Maxmumlkelhood estmaton of rcan dstrbuton parameters, IEEE Transactons on Medcal Imagng, vol. 8, no. 3, pp , 998.

Acoustic Camera Image Mosaicing and Super-resolution

Acoustic Camera Image Mosaicing and Super-resolution Acoustc Camera Image Mosacng and Super-resoluton K. Km Insttute for Bran and Neural Systems Brown Unv. Box 1843 Provdence RI 091, USA ko@brown.edu N. Nerett Insttute for Bran and Neural Systems Brown Unv.

More information

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,

More information

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,

More information

An efficient method to build panoramic image mosaics

An efficient method to build panoramic image mosaics An effcent method to buld panoramc mage mosacs Pattern Recognton Letters vol. 4 003 Dae-Hyun Km Yong-In Yoon Jong-Soo Cho School of Electrcal Engneerng and Computer Scence Kyungpook Natonal Unv. Abstract

More information

MOTION PANORAMA CONSTRUCTION FROM STREAMING VIDEO FOR POWER- CONSTRAINED MOBILE MULTIMEDIA ENVIRONMENTS XUNYU PAN

MOTION PANORAMA CONSTRUCTION FROM STREAMING VIDEO FOR POWER- CONSTRAINED MOBILE MULTIMEDIA ENVIRONMENTS XUNYU PAN MOTION PANORAMA CONSTRUCTION FROM STREAMING VIDEO FOR POWER- CONSTRAINED MOBILE MULTIMEDIA ENVIRONMENTS by XUNYU PAN (Under the Drecton of Suchendra M. Bhandarkar) ABSTRACT In modern tmes, more and more

More information

Parallelism for Nested Loops with Non-uniform and Flow Dependences

Parallelism for Nested Loops with Non-uniform and Flow Dependences Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr

More information

Corner-Based Image Alignment using Pyramid Structure with Gradient Vector Similarity

Corner-Based Image Alignment using Pyramid Structure with Gradient Vector Similarity Journal of Sgnal and Informaton Processng, 013, 4, 114-119 do:10.436/jsp.013.43b00 Publshed Onlne August 013 (http://www.scrp.org/journal/jsp) Corner-Based Image Algnment usng Pyramd Structure wth Gradent

More information

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth

More information

Image Alignment CSC 767

Image Alignment CSC 767 Image Algnment CSC 767 Image algnment Image from http://graphcs.cs.cmu.edu/courses/15-463/2010_fall/ Image algnment: Applcatons Panorama sttchng Image algnment: Applcatons Recognton of object nstances

More information

An Image Fusion Approach Based on Segmentation Region

An Image Fusion Approach Based on Segmentation Region Rong Wang, L-Qun Gao, Shu Yang, Yu-Hua Cha, and Yan-Chun Lu An Image Fuson Approach Based On Segmentaton Regon An Image Fuson Approach Based on Segmentaton Regon Rong Wang, L-Qun Gao, Shu Yang 3, Yu-Hua

More information

Range images. Range image registration. Examples of sampling patterns. Range images and range surfaces

Range images. Range image registration. Examples of sampling patterns. Range images and range surfaces Range mages For many structured lght scanners, the range data forms a hghly regular pattern known as a range mage. he samplng pattern s determned by the specfc scanner. Range mage regstraton 1 Examples

More information

Computer Animation and Visualisation. Lecture 4. Rigging / Skinning

Computer Animation and Visualisation. Lecture 4. Rigging / Skinning Computer Anmaton and Vsualsaton Lecture 4. Rggng / Sknnng Taku Komura Overvew Sknnng / Rggng Background knowledge Lnear Blendng How to decde weghts? Example-based Method Anatomcal models Sknnng Assume

More information

S1 Note. Basis functions.

S1 Note. Basis functions. S1 Note. Bass functons. Contents Types of bass functons...1 The Fourer bass...2 B-splne bass...3 Power and type I error rates wth dfferent numbers of bass functons...4 Table S1. Smulaton results of type

More information

CS 534: Computer Vision Model Fitting

CS 534: Computer Vision Model Fitting CS 534: Computer Vson Model Fttng Sprng 004 Ahmed Elgammal Dept of Computer Scence CS 534 Model Fttng - 1 Outlnes Model fttng s mportant Least-squares fttng Maxmum lkelhood estmaton MAP estmaton Robust

More information

A Robust Method for Estimating the Fundamental Matrix

A Robust Method for Estimating the Fundamental Matrix Proc. VIIth Dgtal Image Computng: Technques and Applcatons, Sun C., Talbot H., Ourseln S. and Adraansen T. (Eds.), 0- Dec. 003, Sydney A Robust Method for Estmatng the Fundamental Matrx C.L. Feng and Y.S.

More information

Image Representation & Visualization Basic Imaging Algorithms Shape Representation and Analysis. outline

Image Representation & Visualization Basic Imaging Algorithms Shape Representation and Analysis. outline mage Vsualzaton mage Vsualzaton mage Representaton & Vsualzaton Basc magng Algorthms Shape Representaton and Analyss outlne mage Representaton & Vsualzaton Basc magng Algorthms Shape Representaton and

More information

TN348: Openlab Module - Colocalization

TN348: Openlab Module - Colocalization TN348: Openlab Module - Colocalzaton Topc The Colocalzaton module provdes the faclty to vsualze and quantfy colocalzaton between pars of mages. The Colocalzaton wndow contans a prevew of the two mages

More information

A Binarization Algorithm specialized on Document Images and Photos

A Binarization Algorithm specialized on Document Images and Photos A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a

More information

3D vector computer graphics

3D vector computer graphics 3D vector computer graphcs Paolo Varagnolo: freelance engneer Padova Aprl 2016 Prvate Practce ----------------------------------- 1. Introducton Vector 3D model representaton n computer graphcs requres

More information

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching A Fast Vsual Trackng Algorthm Based on Crcle Pxels Matchng Zhqang Hou hou_zhq@sohu.com Chongzhao Han czhan@mal.xjtu.edu.cn Ln Zheng Abstract: A fast vsual trackng algorthm based on crcle pxels matchng

More information

R s s f. m y s. SPH3UW Unit 7.3 Spherical Concave Mirrors Page 1 of 12. Notes

R s s f. m y s. SPH3UW Unit 7.3 Spherical Concave Mirrors Page 1 of 12. Notes SPH3UW Unt 7.3 Sphercal Concave Mrrors Page 1 of 1 Notes Physcs Tool box Concave Mrror If the reflectng surface takes place on the nner surface of the sphercal shape so that the centre of the mrror bulges

More information

Fitting & Matching. Lecture 4 Prof. Bregler. Slides from: S. Lazebnik, S. Seitz, M. Pollefeys, A. Effros.

Fitting & Matching. Lecture 4 Prof. Bregler. Slides from: S. Lazebnik, S. Seitz, M. Pollefeys, A. Effros. Fttng & Matchng Lecture 4 Prof. Bregler Sldes from: S. Lazebnk, S. Setz, M. Pollefeys, A. Effros. How do we buld panorama? We need to match (algn) mages Matchng wth Features Detect feature ponts n both

More information

THE PULL-PUSH ALGORITHM REVISITED

THE PULL-PUSH ALGORITHM REVISITED THE PULL-PUSH ALGORITHM REVISITED Improvements, Computaton of Pont Denstes, and GPU Implementaton Martn Kraus Computer Graphcs & Vsualzaton Group, Technsche Unverstät München, Boltzmannstraße 3, 85748

More information

Positive Semi-definite Programming Localization in Wireless Sensor Networks

Positive Semi-definite Programming Localization in Wireless Sensor Networks Postve Sem-defnte Programmng Localzaton n Wreless Sensor etworks Shengdong Xe 1,, Jn Wang, Aqun Hu 1, Yunl Gu, Jang Xu, 1 School of Informaton Scence and Engneerng, Southeast Unversty, 10096, anjng Computer

More information

METRIC ALIGNMENT OF LASER RANGE SCANS AND CALIBRATED IMAGES USING LINEAR STRUCTURES

METRIC ALIGNMENT OF LASER RANGE SCANS AND CALIBRATED IMAGES USING LINEAR STRUCTURES METRIC ALIGNMENT OF LASER RANGE SCANS AND CALIBRATED IMAGES USING LINEAR STRUCTURES Lorenzo Sorg CIRA the Italan Aerospace Research Centre Computer Vson and Vrtual Realty Lab. Outlne Work goal Work motvaton

More information

Fuzzy Filtering Algorithms for Image Processing: Performance Evaluation of Various Approaches

Fuzzy Filtering Algorithms for Image Processing: Performance Evaluation of Various Approaches Proceedngs of the Internatonal Conference on Cognton and Recognton Fuzzy Flterng Algorthms for Image Processng: Performance Evaluaton of Varous Approaches Rajoo Pandey and Umesh Ghanekar Department of

More information

User Authentication Based On Behavioral Mouse Dynamics Biometrics

User Authentication Based On Behavioral Mouse Dynamics Biometrics User Authentcaton Based On Behavoral Mouse Dynamcs Bometrcs Chee-Hyung Yoon Danel Donghyun Km Department of Computer Scence Department of Computer Scence Stanford Unversty Stanford Unversty Stanford, CA

More information

Outline. Type of Machine Learning. Examples of Application. Unsupervised Learning

Outline. Type of Machine Learning. Examples of Application. Unsupervised Learning Outlne Artfcal Intellgence and ts applcatons Lecture 8 Unsupervsed Learnng Professor Danel Yeung danyeung@eee.org Dr. Patrck Chan patrckchan@eee.org South Chna Unversty of Technology, Chna Introducton

More information

A high precision collaborative vision measurement of gear chamfering profile

A high precision collaborative vision measurement of gear chamfering profile Internatonal Conference on Advances n Mechancal Engneerng and Industral Informatcs (AMEII 05) A hgh precson collaboratve vson measurement of gear chamferng profle Conglng Zhou, a, Zengpu Xu, b, Chunmng

More information

Edge Detection in Noisy Images Using the Support Vector Machines

Edge Detection in Noisy Images Using the Support Vector Machines Edge Detecton n Nosy Images Usng the Support Vector Machnes Hlaro Gómez-Moreno, Saturnno Maldonado-Bascón, Francsco López-Ferreras Sgnal Theory and Communcatons Department. Unversty of Alcalá Crta. Madrd-Barcelona

More information

Steps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and. Accessibility (Gravity with Competition) Indices

Steps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and. Accessibility (Gravity with Competition) Indices Steps for Computng the Dssmlarty, Entropy, Herfndahl-Hrschman and Accessblty (Gravty wth Competton) Indces I. Dssmlarty Index Measurement: The followng formula can be used to measure the evenness between

More information

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points;

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points; Subspace clusterng Clusterng Fundamental to all clusterng technques s the choce of dstance measure between data ponts; D q ( ) ( ) 2 x x = x x, j k = 1 k jk Squared Eucldean dstance Assumpton: All features

More information

Some Tutorial about the Project. Computer Graphics

Some Tutorial about the Project. Computer Graphics Some Tutoral about the Project Lecture 6 Rastersaton, Antalasng, Texture Mappng, I have already covered all the topcs needed to fnsh the 1 st practcal Today, I wll brefly explan how to start workng on

More information

Multi-stable Perception. Necker Cube

Multi-stable Perception. Necker Cube Mult-stable Percepton Necker Cube Spnnng dancer lluson, Nobuuk Kaahara Fttng and Algnment Computer Vson Szelsk 6.1 James Has Acknowledgment: Man sldes from Derek Hoem, Lana Lazebnk, and Grauman&Lebe 2008

More information

y and the total sum of

y and the total sum of Lnear regresson Testng for non-lnearty In analytcal chemstry, lnear regresson s commonly used n the constructon of calbraton functons requred for analytcal technques such as gas chromatography, atomc absorpton

More information

Structure from Motion

Structure from Motion Structure from Moton Structure from Moton For now, statc scene and movng camera Equvalentl, rgdl movng scene and statc camera Lmtng case of stereo wth man cameras Lmtng case of multvew camera calbraton

More information

Unsupervised Learning and Clustering

Unsupervised Learning and Clustering Unsupervsed Learnng and Clusterng Why consder unlabeled samples?. Collectng and labelng large set of samples s costly Gettng recorded speech s free, labelng s tme consumng 2. Classfer could be desgned

More information

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data A Fast Content-Based Multmeda Retreval Technque Usng Compressed Data Borko Furht and Pornvt Saksobhavvat NSF Multmeda Laboratory Florda Atlantc Unversty, Boca Raton, Florda 3343 ABSTRACT In ths paper,

More information

Image Mosaicing of Noisy Acoustic Camera. Images

Image Mosaicing of Noisy Acoustic Camera. Images Image Mosaicing of Noisy Acoustic Camera Images Kio Kim, Nathan Intrator and Nicola Neretti May 21, 2004 Abstract This paper presents an algorithm for image registration and mosaicing on underwater sonar

More information

Hermite Splines in Lie Groups as Products of Geodesics

Hermite Splines in Lie Groups as Products of Geodesics Hermte Splnes n Le Groups as Products of Geodescs Ethan Eade Updated May 28, 2017 1 Introducton 1.1 Goal Ths document defnes a curve n the Le group G parametrzed by tme and by structural parameters n the

More information

Lobachevsky State University of Nizhni Novgorod. Polyhedron. Quick Start Guide

Lobachevsky State University of Nizhni Novgorod. Polyhedron. Quick Start Guide Lobachevsky State Unversty of Nzhn Novgorod Polyhedron Quck Start Gude Nzhn Novgorod 2016 Contents Specfcaton of Polyhedron software... 3 Theoretcal background... 4 1. Interface of Polyhedron... 6 1.1.

More information

Prof. Feng Liu. Spring /24/2017

Prof. Feng Liu. Spring /24/2017 Prof. Feng Lu Sprng 2017 ttp://www.cs.pd.edu/~flu/courses/cs510/ 05/24/2017 Last me Compostng and Mattng 2 oday Vdeo Stablzaton Vdeo stablzaton ppelne 3 Orson Welles, ouc of Evl, 1958 4 Images courtesy

More information

Unsupervised Learning

Unsupervised Learning Pattern Recognton Lecture 8 Outlne Introducton Unsupervsed Learnng Parametrc VS Non-Parametrc Approach Mxture of Denstes Maxmum-Lkelhood Estmates Clusterng Prof. Danel Yeung School of Computer Scence and

More information

Real-time Joint Tracking of a Hand Manipulating an Object from RGB-D Input

Real-time Joint Tracking of a Hand Manipulating an Object from RGB-D Input Real-tme Jont Tracng of a Hand Manpulatng an Object from RGB-D Input Srnath Srdhar 1 Franzsa Mueller 1 Mchael Zollhöfer 1 Dan Casas 1 Antt Oulasvrta 2 Chrstan Theobalt 1 1 Max Planc Insttute for Informatcs

More information

A Super-resolution Algorithm Based on SURF and POCS for 3D Bionics PTZ

A Super-resolution Algorithm Based on SURF and POCS for 3D Bionics PTZ Sensors & Transducers 204 by IFSA Publshng, S. L. http://www.sensorsportal.com A Super-resoluton Algorthm Based on SURF and POCS for 3D Boncs PTZ Hengyu LI, Jqng CHEN, Shaorong XIE and Jun LUO School of

More information

Image warping and stitching May 5 th, 2015

Image warping and stitching May 5 th, 2015 Image warpng and sttchng Ma 5 th, 2015 Yong Jae Lee UC Davs PS2 due net Frda Announcements 2 Last tme Interactve segmentaton Feature-based algnment 2D transformatons Affne ft RANSAC 3 1 Algnment problem

More information

Image Fusion With a Dental Panoramic X-ray Image and Face Image Acquired With a KINECT

Image Fusion With a Dental Panoramic X-ray Image and Face Image Acquired With a KINECT Image Fuson Wth a Dental Panoramc X-ray Image and Face Image Acqured Wth a KINECT Kohe Kawa* 1, Koch Ogawa* 1, Aktosh Katumata* 2 * 1 Graduate School of Engneerng, Hose Unversty * 2 School of Dentstry,

More information

MOTION BLUR ESTIMATION AT CORNERS

MOTION BLUR ESTIMATION AT CORNERS Gacomo Boracch and Vncenzo Caglot Dpartmento d Elettronca e Informazone, Poltecnco d Mlano, Va Ponzo, 34/5-20133 MILANO boracch@elet.polm.t, caglot@elet.polm.t Keywords: Abstract: Pont Spread Functon Parameter

More information

Ecient Computation of the Most Probable Motion from Fuzzy. Moshe Ben-Ezra Shmuel Peleg Michael Werman. The Hebrew University of Jerusalem

Ecient Computation of the Most Probable Motion from Fuzzy. Moshe Ben-Ezra Shmuel Peleg Michael Werman. The Hebrew University of Jerusalem Ecent Computaton of the Most Probable Moton from Fuzzy Correspondences Moshe Ben-Ezra Shmuel Peleg Mchael Werman Insttute of Computer Scence The Hebrew Unversty of Jerusalem 91904 Jerusalem, Israel Emal:

More information

Feature-Area Optimization: A Novel SAR Image Registration Method

Feature-Area Optimization: A Novel SAR Image Registration Method Feature-Area Optmzaton: A Novel SAR Image Regstraton Method Fuqang Lu, Fukun B, Lang Chen, Hao Sh and We Lu Abstract Ths letter proposes a synthetc aperture radar (SAR) mage regstraton method named Feature-Area

More information

Lecture 5: Multilayer Perceptrons

Lecture 5: Multilayer Perceptrons Lecture 5: Multlayer Perceptrons Roger Grosse 1 Introducton So far, we ve only talked about lnear models: lnear regresson and lnear bnary classfers. We noted that there are functons that can t be represented

More information

Outline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1

Outline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1 4/14/011 Outlne Dscrmnatve classfers for mage recognton Wednesday, Aprl 13 Krsten Grauman UT-Austn Last tme: wndow-based generc obect detecton basc ppelne face detecton wth boostng as case study Today:

More information

Feature-based image registration using the shape context

Feature-based image registration using the shape context Feature-based mage regstraton usng the shape context LEI HUANG *, ZHEN LI Center for Earth Observaton and Dgtal Earth, Chnese Academy of Scences, Bejng, 100012, Chna Graduate Unversty of Chnese Academy

More information

The Research of Ellipse Parameter Fitting Algorithm of Ultrasonic Imaging Logging in the Casing Hole

The Research of Ellipse Parameter Fitting Algorithm of Ultrasonic Imaging Logging in the Casing Hole Appled Mathematcs, 04, 5, 37-3 Publshed Onlne May 04 n ScRes. http://www.scrp.org/journal/am http://dx.do.org/0.436/am.04.584 The Research of Ellpse Parameter Fttng Algorthm of Ultrasonc Imagng Loggng

More information

Inverse-Polar Ray Projection for Recovering Projective Transformations

Inverse-Polar Ray Projection for Recovering Projective Transformations nverse-polar Ray Projecton for Recoverng Projectve Transformatons Yun Zhang The Center for Advanced Computer Studes Unversty of Lousana at Lafayette yxz646@lousana.edu Henry Chu The Center for Advanced

More information

n others; multple brghtness values n one mage may map to a sngle brghtness value n the other mage, and vce versa. In other words, the two mages are us

n others; multple brghtness values n one mage may map to a sngle brghtness value n the other mage, and vce versa. In other words, the two mages are us Robust Mult-Sensor Image Algnment Mchal Iran Dept. of Appled Math and CS The Wezmann Insttute of Scence 76100 Rehovot, Israel P. Anandan Mcrosoft Corporaton One Mcrosoft Way Redmond, WA 98052, USA Abstract

More information

Real-time Motion Capture System Using One Video Camera Based on Color and Edge Distribution

Real-time Motion Capture System Using One Video Camera Based on Color and Edge Distribution Real-tme Moton Capture System Usng One Vdeo Camera Based on Color and Edge Dstrbuton YOSHIAKI AKAZAWA, YOSHIHIRO OKADA, AND KOICHI NIIJIMA Graduate School of Informaton Scence and Electrcal Engneerng,

More information

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Proceedngs of the Wnter Smulaton Conference M E Kuhl, N M Steger, F B Armstrong, and J A Jones, eds A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Mark W Brantley Chun-Hung

More information

A Range Image Refinement Technique for Multi-view 3D Model Reconstruction

A Range Image Refinement Technique for Multi-view 3D Model Reconstruction A Range Image Refnement Technque for Mult-vew 3D Model Reconstructon Soon-Yong Park and Mural Subbarao Electrcal and Computer Engneerng State Unversty of New York at Stony Brook, USA E-mal: parksy@ece.sunysb.edu

More information

Multi-view 3D Position Estimation of Sports Players

Multi-view 3D Position Estimation of Sports Players Mult-vew 3D Poston Estmaton of Sports Players Robbe Vos and Wlle Brnk Appled Mathematcs Department of Mathematcal Scences Unversty of Stellenbosch, South Afrca Emal: vosrobbe@gmal.com Abstract The problem

More information

Dynamic wetting property investigation of AFM tips in micro/nanoscale

Dynamic wetting property investigation of AFM tips in micro/nanoscale Dynamc wettng property nvestgaton of AFM tps n mcro/nanoscale The wettng propertes of AFM probe tps are of concern n AFM tp related force measurement, fabrcaton, and manpulaton technques, such as dp-pen

More information

UAV global pose estimation by matching forward-looking aerial images with satellite images

UAV global pose estimation by matching forward-looking aerial images with satellite images The 2009 IEEE/RSJ Internatonal Conference on Intellgent Robots and Systems October -5, 2009 St. Lous, USA UAV global pose estmaton by matchng forward-lookng aeral mages wth satellte mages Kl-Ho Son, Youngbae

More information

Lecture 13: High-dimensional Images

Lecture 13: High-dimensional Images Lec : Hgh-dmensonal Images Grayscale Images Lecture : Hgh-dmensonal Images Math 90 Prof. Todd Wttman The Ctadel A grayscale mage s an nteger-valued D matrx. An 8-bt mage takes on values between 0 and 55.

More information

Skew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach

Skew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach Angle Estmaton and Correcton of Hand Wrtten, Textual and Large areas of Non-Textual Document Images: A Novel Approach D.R.Ramesh Babu Pyush M Kumat Mahesh D Dhannawat PES Insttute of Technology Research

More information

3D Modeling Using Multi-View Images. Jinjin Li. A Thesis Presented in Partial Fulfillment of the Requirements for the Degree Master of Science

3D Modeling Using Multi-View Images. Jinjin Li. A Thesis Presented in Partial Fulfillment of the Requirements for the Degree Master of Science 3D Modelng Usng Mult-Vew Images by Jnjn L A Thess Presented n Partal Fulfllment of the Requrements for the Degree Master of Scence Approved August by the Graduate Supervsory Commttee: Lna J. Karam, Char

More information

AUTOMATIC IMAGE REGISTRATION OF MULTI-ANGLE IMAGERY FOR CHRIS/PROBA

AUTOMATIC IMAGE REGISTRATION OF MULTI-ANGLE IMAGERY FOR CHRIS/PROBA AUTOMATIC IMAGE REGISTRATION OF MULTI-ANGLE IMAGERY FOR CHRIS/PROBA J. Ma *, J.C.-W. Chan, F. Canters Cartography and GIS Research Group, Department of Geography, Vrje Unverstet Brussel, Plenlaan, 050

More information

Mathematics 256 a course in differential equations for engineering students

Mathematics 256 a course in differential equations for engineering students Mathematcs 56 a course n dfferental equatons for engneerng students Chapter 5. More effcent methods of numercal soluton Euler s method s qute neffcent. Because the error s essentally proportonal to the

More information

Fitting: Deformable contours April 26 th, 2018

Fitting: Deformable contours April 26 th, 2018 4/6/08 Fttng: Deformable contours Aprl 6 th, 08 Yong Jae Lee UC Davs Recap so far: Groupng and Fttng Goal: move from array of pxel values (or flter outputs) to a collecton of regons, objects, and shapes.

More information

ESTIMATION OF INTERIOR ORIENTATION AND ECCENTRICITY PARAMETERS OF A HYBRID IMAGING AND LASER SCANNING SENSOR

ESTIMATION OF INTERIOR ORIENTATION AND ECCENTRICITY PARAMETERS OF A HYBRID IMAGING AND LASER SCANNING SENSOR ESTIMATION OF INTERIOR ORIENTATION AND ECCENTRICITY PARAMETERS OF A HYBRID IMAGING AND LASER SCANNING SENSOR A. Wendt a, C. Dold b a Insttute for Appled Photogrammetry and Geonformatcs, Unversty of Appled

More information

Feature Reduction and Selection

Feature Reduction and Selection Feature Reducton and Selecton Dr. Shuang LIANG School of Software Engneerng TongJ Unversty Fall, 2012 Today s Topcs Introducton Problems of Dmensonalty Feature Reducton Statstc methods Prncpal Components

More information

Proper Choice of Data Used for the Estimation of Datum Transformation Parameters

Proper Choice of Data Used for the Estimation of Datum Transformation Parameters Proper Choce of Data Used for the Estmaton of Datum Transformaton Parameters Hakan S. KUTOGLU, Turkey Key words: Coordnate systems; transformaton; estmaton, relablty. SUMMARY Advances n technologes and

More information

Optimal Scheduling of Capture Times in a Multiple Capture Imaging System

Optimal Scheduling of Capture Times in a Multiple Capture Imaging System Optmal Schedulng of Capture Tmes n a Multple Capture Imagng System Tng Chen and Abbas El Gamal Informaton Systems Laboratory Department of Electrcal Engneerng Stanford Unversty Stanford, Calforna 9435,

More information

Fitting and Alignment

Fitting and Alignment Fttng and Algnment Computer Vson Ja-Bn Huang, Vrgna Tech Many sldes from S. Lazebnk and D. Hoem Admnstratve Stuffs HW 1 Competton: Edge Detecton Submsson lnk HW 2 wll be posted tonght Due Oct 09 (Mon)

More information

MAPPING CROP STATUS FROM AN UNMANNED AERIAL VEHICLE FOR PRECISION AGRICULTURE APPLICATIONS

MAPPING CROP STATUS FROM AN UNMANNED AERIAL VEHICLE FOR PRECISION AGRICULTURE APPLICATIONS XXII ISPRS Congress, 25 August 01 September 2012, elbourne, Australa APPING CROP STATUS FRO AN UNANNED AERIAL VEHICLE FOR PRECISION AGRICULTURE APPLICATIONS T. Guo, T. Kujra, T. Watanabe Htach, Ltd., Central

More information

Face Tracking Using Motion-Guided Dynamic Template Matching

Face Tracking Using Motion-Guided Dynamic Template Matching ACCV2002: The 5th Asan Conference on Computer Vson, 23--25 January 2002, Melbourne, Australa. Face Trackng Usng Moton-Guded Dynamc Template Matchng Lang Wang, Tenu Tan, Wemng Hu atonal Laboratory of Pattern

More information

Registration of Expressions Data using a 3D Morphable Model

Registration of Expressions Data using a 3D Morphable Model Regstraton of Expressons Data usng a 3D Morphable Model Curzo Basso DISI, Unverstà d Genova, Italy Emal: curzo.basso@ds.unge.t Thomas Vetter Departement Informatk, Unverstät Basel, Swtzerland Emal: thomas.vetter@unbas.ch

More information

Reducing Frame Rate for Object Tracking

Reducing Frame Rate for Object Tracking Reducng Frame Rate for Object Trackng Pavel Korshunov 1 and We Tsang Oo 2 1 Natonal Unversty of Sngapore, Sngapore 11977, pavelkor@comp.nus.edu.sg 2 Natonal Unversty of Sngapore, Sngapore 11977, oowt@comp.nus.edu.sg

More information

POTENTIAL OF DENSE MATCHING FOR THE GENERATION OF HIGH QUALITY DIGITAL ELEVATION MODELS

POTENTIAL OF DENSE MATCHING FOR THE GENERATION OF HIGH QUALITY DIGITAL ELEVATION MODELS POTENTIAL OF DENSE MATCHING FOR THE GENERATION OF HIGH QUALITY DIGITAL ELEVATION MODELS Mathas Rothermel, Norbert Haala Insttute for Photogrammetry (fp), Unversty of Stuttgart, Geschwster-Scholl-Str. 24D,

More information

FEATURE EXTRACTION. Dr. K.Vijayarekha. Associate Dean School of Electrical and Electronics Engineering SASTRA University, Thanjavur

FEATURE EXTRACTION. Dr. K.Vijayarekha. Associate Dean School of Electrical and Electronics Engineering SASTRA University, Thanjavur FEATURE EXTRACTION Dr. K.Vjayarekha Assocate Dean School of Electrcal and Electroncs Engneerng SASTRA Unversty, Thanjavur613 41 Jont Intatve of IITs and IISc Funded by MHRD Page 1 of 8 Table of Contents

More information

Face Recognition University at Buffalo CSE666 Lecture Slides Resources:

Face Recognition University at Buffalo CSE666 Lecture Slides Resources: Face Recognton Unversty at Buffalo CSE666 Lecture Sldes Resources: http://www.face-rec.org/algorthms/ Overvew of face recognton algorthms Correlaton - Pxel based correspondence between two face mages Structural

More information

Distance Calculation from Single Optical Image

Distance Calculation from Single Optical Image 17 Internatonal Conference on Mathematcs, Modellng and Smulaton Technologes and Applcatons (MMSTA 17) ISBN: 978-1-6595-53-8 Dstance Calculaton from Sngle Optcal Image Xao-yng DUAN 1,, Yang-je WEI 1,,*

More information

2x x l. Module 3: Element Properties Lecture 4: Lagrange and Serendipity Elements

2x x l. Module 3: Element Properties Lecture 4: Lagrange and Serendipity Elements Module 3: Element Propertes Lecture : Lagrange and Serendpty Elements 5 In last lecture note, the nterpolaton functons are derved on the bass of assumed polynomal from Pascal s trangle for the fled varable.

More information

A Volumetric Approach for Interactive 3D Modeling

A Volumetric Approach for Interactive 3D Modeling A Volumetrc Approach for Interactve 3D Modelng Dragan Tubć Patrck Hébert Computer Vson and Systems Laboratory Laval Unversty, Ste-Foy, Québec, Canada, G1K 7P4 Dens Laurendeau E-mal: (tdragan, hebert, laurendeau)@gel.ulaval.ca

More information

A NEW IMPLEMENTATION OF THE ICP ALGORITHM FOR 3D SURFACE REGISTRATION USING A COMPREHENSIVE LOOK UP MATRIX

A NEW IMPLEMENTATION OF THE ICP ALGORITHM FOR 3D SURFACE REGISTRATION USING A COMPREHENSIVE LOOK UP MATRIX A NEW IMPLEMENTATION OF THE ICP ALGORITHM FOR 3D SURFACE REGISTRATION USING A COMPREHENSIVE LOOK UP MATRIX A. Almhde, C. Léger, M. Derche 2 and R. Lédée Laboratory of Electroncs, Sgnals and Images (LESI),

More information

Six-Band HDTV Camera System for Color Reproduction Based on Spectral Information

Six-Band HDTV Camera System for Color Reproduction Based on Spectral Information IS&T's 23 PICS Conference Sx-Band HDTV Camera System for Color Reproducton Based on Spectral Informaton Kenro Ohsawa )4), Hroyuk Fukuda ), Takeyuk Ajto 2),Yasuhro Komya 2), Hdeak Hanesh 3), Masahro Yamaguch

More information

A Comparison and Evaluation of Three Different Pose Estimation Algorithms In Detecting Low Texture Manufactured Objects

A Comparison and Evaluation of Three Different Pose Estimation Algorithms In Detecting Low Texture Manufactured Objects Clemson Unversty TgerPrnts All Theses Theses 12-2011 A Comparson and Evaluaton of Three Dfferent Pose Estmaton Algorthms In Detectng Low Texture Manufactured Objects Robert Krener Clemson Unversty, rkrene@clemson.edu

More information

Optimal Combination of Stereo Camera Calibration from Arbitrary Stereo Images.

Optimal Combination of Stereo Camera Calibration from Arbitrary Stereo Images. Tna Memo No. 1991-002 Image and Vson Computng, 9(1), 27-32, 1990. Optmal Combnaton of Stereo Camera Calbraton from Arbtrary Stereo Images. N.A.Thacker and J.E.W.Mayhew. Last updated 6 / 9 / 2005 Imagng

More information

Joint Example-based Depth Map Super-Resolution

Joint Example-based Depth Map Super-Resolution Jont Example-based Depth Map Super-Resoluton Yanje L 1, Tanfan Xue,3, Lfeng Sun 1, Janzhuang Lu,3,4 1 Informaton Scence and Technology Department, Tsnghua Unversty, Bejng, Chna Department of Informaton

More information

Support Vector Machines

Support Vector Machines /9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.

More information

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour 6.854 Advanced Algorthms Petar Maymounkov Problem Set 11 (November 23, 2005) Wth: Benjamn Rossman, Oren Wemann, and Pouya Kheradpour Problem 1. We reduce vertex cover to MAX-SAT wth weghts, such that the

More information

A Background Subtraction for a Vision-based User Interface *

A Background Subtraction for a Vision-based User Interface * A Background Subtracton for a Vson-based User Interface * Dongpyo Hong and Woontack Woo KJIST U-VR Lab. {dhon wwoo}@kjst.ac.kr Abstract In ths paper, we propose a robust and effcent background subtracton

More information

Image-based Motion Stabilization for Maritime Surveillance

Image-based Motion Stabilization for Maritime Surveillance Image-based Moton Stablzaton for Martme Survellance Danel D. Morrs, Bran R. Colonna and Franln D. Snyder General Dynamcs Robotc Systems, 151 Ardmore Blvd, Pttsburgh, PA 15221 * ABSTRACT Robust mage-based

More information

Line-based Camera Movement Estimation by Using Parallel Lines in Omnidirectional Video

Line-based Camera Movement Estimation by Using Parallel Lines in Omnidirectional Video 01 IEEE Internatonal Conference on Robotcs and Automaton RverCentre, Sant Paul, Mnnesota, USA May 14-18, 01 Lne-based Camera Movement Estmaton by Usng Parallel Lnes n Omndrectonal Vdeo Ryosuke kawansh,

More information

Complex Filtering and Integration via Sampling

Complex Filtering and Integration via Sampling Overvew Complex Flterng and Integraton va Samplng Sgnal processng Sample then flter (remove alases) then resample onunform samplng: jtterng and Posson dsk Statstcs Monte Carlo ntegraton and probablty theory

More information

An Entropy-Based Approach to Integrated Information Needs Assessment

An Entropy-Based Approach to Integrated Information Needs Assessment Dstrbuton Statement A: Approved for publc release; dstrbuton s unlmted. An Entropy-Based Approach to ntegrated nformaton Needs Assessment June 8, 2004 Wllam J. Farrell Lockheed Martn Advanced Technology

More information

PERFORMANCE EVALUATION FOR SCENE MATCHING ALGORITHMS BY SVM

PERFORMANCE EVALUATION FOR SCENE MATCHING ALGORITHMS BY SVM PERFORMACE EVALUAIO FOR SCEE MACHIG ALGORIHMS BY SVM Zhaohu Yang a, b, *, Yngyng Chen a, Shaomng Zhang a a he Research Center of Remote Sensng and Geomatc, ongj Unversty, Shangha 200092, Chna - yzhac@63.com

More information

Accounting for the Use of Different Length Scale Factors in x, y and z Directions

Accounting for the Use of Different Length Scale Factors in x, y and z Directions 1 Accountng for the Use of Dfferent Length Scale Factors n x, y and z Drectons Taha Soch (taha.soch@kcl.ac.uk) Imagng Scences & Bomedcal Engneerng, Kng s College London, The Rayne Insttute, St Thomas Hosptal,

More information

Palmprint Feature Extraction Using 2-D Gabor Filters

Palmprint Feature Extraction Using 2-D Gabor Filters Palmprnt Feature Extracton Usng 2-D Gabor Flters Wa Kn Kong Davd Zhang and Wenxn L Bometrcs Research Centre Department of Computng The Hong Kong Polytechnc Unversty Kowloon Hong Kong Correspondng author:

More information

Implementation of a Dynamic Image-Based Rendering System

Implementation of a Dynamic Image-Based Rendering System Implementaton of a Dynamc Image-Based Renderng System Nklas Bakos, Claes Järvman and Mark Ollla 3 Norrköpng Vsualzaton and Interacton Studo Lnköpng Unversty Abstract Work n dynamc mage based renderng has

More information

Hierarchical clustering for gene expression data analysis

Hierarchical clustering for gene expression data analysis Herarchcal clusterng for gene expresson data analyss Gorgo Valentn e-mal: valentn@ds.unm.t Clusterng of Mcroarray Data. Clusterng of gene expresson profles (rows) => dscovery of co-regulated and functonally

More information