IMPLEMENTATION OF COMPUTER VISION TECHNIQUES USING OPENCV

Transcription

1 IMPLEMENTATION OF COMPUTER VISION TECHNIQUES USING OPENCV Anu Suneja Assistant Professor M. M. University Mullana, Haryana, India From last few years, the thrust for automation and simulation tools in multiple applications are in escalation to solve the research problems. Numerous domains including network applications, cyber security, digital image processing, cloud computing, data mining, machine learning, computer vision and many others are rapidly getting the software products developed in multiple languages for implementation of algorithms or simulation of the techniques to get the accurate and specific results without using core programming. Classically, core programming languages and research oriented tools including C, C++, Java, MATLAB, SciLab and many other suites are used for the automation of paradigm algorithms. Still, lots of domains are there which are so complex in which core programming is very time consuming and despite of the core coding, getting the specific and accurate results is an issue. Computer Vision is one of such domain that is having lots of algorithms which are very time consuming and complex in terms of core programming. The core areas in computer vision includes 2D and 3D features, Egomotion estimation, Facial recognition system, Gesture recognition, Human computer interaction (HCI), Mobile robotics, Motion understanding, Object identification, Segmentation Recognition, Stereopsis Stereo vision: depth perception from 2

2 cameras, Structure from motion (SFM), Motion tracking, Augmented reality, pattern recognition, scene reconstruction, decision making and many others. However, the computer vision algorithms can be implemented / simulated in MATLAB / SciLab / VC++, still the rapid implementation will be the issue until we have a specific tool for computer vision techniques. Here, we will highlight a specific open source tool OPENCV (Open Source Computer Vision Library) for the implementation of computer vision algorithms and technique. OPENCV is widely used in research laboratories and industry to achieve the explicit and efficient results. OpenCV [ is an open source library of the functions primarily related to the real-time computer vision techniques. There is download statistics of more than 40,500 downloads per week from SourceForge.net. OpenCV is developed by Intel and currently supported by Willow Garage and Itseez. The software suite is free for use, cross-platform and under open source BSD license. The software library focuses very efficiently on real-time image processing and computer vision algorithms. The fame of OpenCV can be analyzed from the fact that it has more than 47 thousand people of user community and estimated number of downloads exceeding 6 million. Along with wellestablished companies like Google, Yahoo, Microsoft, Intel, IBM, Sony, Honda, Toyota that employ the library, there are many startups such as Applied Minds, VideoSurf, and Zeitera, that make extensive use of OpenCV. OpenCV s deployed uses span the range from stitching streetview images together, detecting intrusions in surveillance video in Israel, monitoring mine equipment in China, helping robots navigate and pick up objects at Willow Garage, detection of swimming pool drowning accidents in Europe, running interactive art in Spain and New York, checking runways for debris in Turkey, inspecting labels on products in factories around the world on to rapid face detection in Japan. [Source : There is a Yahoo Groups Forum where users can post questions and discussion at it has about 20,000 members.

3 OpenCV can be executed successfully without any issue on Windows, Android, Maemo, FreeBSD, OpenBSD, ios, BlackBerry, Linux and OS X. The official releases of OpenCV can be obtained from SourceForge. Statistical Machine Learning library used by OpenCV Boosting (meta-algorithm) Decision tree learning Gradient boosting trees Expectation-maximization algorithm k-nearest neighbor algorithm Naive Bayes classifier Artificial neural networks Random forest Support vector machine (SVM) OpenCV is basically written in C++. The primary interface of OpenCV is in C++. Now, there are full interfaces available in Python, Java and MATLAB/OCTAVE. OpenCV was developed for computational efficiency with a strong focus on realtime applications. One of the goals of OpenCV is to offer an easy to use computer vision infrastructure that helps the users to build sophisticated vision applications rapidly. The OpenCV library is having more than 500 functions that span across many areas in computer vision including factory product inspection, medical imaging, security, user interface, camera calibration, stereo vision and robotics. OpenCV Documentation

4 OpenCV s documentation Wiki is more up-to-date than the html pages that ship with OpenCV and it also features additional content as well. The Wiki is located at It includes information on: Instructions on compiling OpenCV using Eclipse IDE Face recognition with OpenCV Video surveillance library Tutorials Camera compatibility Links to the Chinese and the Korean user groups Stereo correspondence View point morphing of cameras 3D tracking in stereo Eigen object (PCA) functions for object recognition Embedded hidden Markov models (HMMs) OpenCV Structure and Content OpenCV is broadly structured into fi ve main components, four of which are shown in Figure 2. The CV component contains the basic image processing and higher-level computer vision algorithms; ML is the machine learning library, which includes many statistical classifi ers and clustering tools. HighGUI contains I/O routines and functions for storing and loading video and images, and CXCore contains the basic data structures and content.

5 Using OpenCV After installation of the OpenCV library, first of all we will execute simple algorithms. for this we have to set up the programming environment. In Visual Studio, we have to create a project and to configure the setup so that (a) the libraries highgui.lib, cxcore.lib, ml.lib, and cv.lib are linked* and (b) the preprocessor will search the OpenCV /opencv/*/include directories for header files. These include directories will typically be named something like C:/program files/opencv/ cv/include, /opencv/cxcore/include, /opencv/ml/include, and /opencv/otherlibs/ highgui. Once we have done this, we will create a new C file for the first program. To Display a Digital Image loaded from the disk OpenCV provides the utilities for accessing a wide array of image file types as well as from video and cameras. Such utilities are part of a toolkit called HighGUI, that is included in the OpenCV package. We will make use of some of these utilities to create a simple program that opens an image and displays it on the screen #include highgui.h int main( int argc, char** argv ){ IplImage* img = cvloadimage( argv[1] ); cvnamedwindow( Example1, CV_WINDOW_AUTOSIZE ); cvshowimage( Example1, img ); cvwaitkey(0);

6 cvreleaseimage( &img ); cvdestroywindow( Example1 ); } When this code is compiled and executed from the command line with a single argument, the program loads the image into memory and displays it on the screen. Then it will wait until the user presses a key, at which time it closes the window and exits. Reading an AVI Video File Execution or playing a video file with OpenCV is as easy as displaying a single picture. #include highgui.h int main( int argc, char** argv ){ cvnamedwindow( Example2, CV_WINDOW_AUTOSIZE ); CvCapture* capture = cvcreatefilecapture( argv[1] ); IplImage* frame; while(1) { frame = cvqueryframe( capture ); if(!frame ) break; cvshowimage( Example2, frame ); char c = cvwaitkey(33); if( c == 27 ) break; } cvreleasecapture( &capture ); cvdestroywindow( Example2 ); } Accessing the Webcam using OpenCV

7 Accessing the webcam or camera device of computer is not very big task in OpenCV. We can simply call open() on a cv::videocapture object (OpenCV's method of accessing your camera device), and pass 0 as the default camera ID number. Some systems have multiple cameras attached or they do not work as default camera 0; so it is common practice to allow the user to pass the desired camera number as a command-line argument, in case they want to try camera 1, 2, or -1, for example. We can also try to set the camera resolution to 640 x 480 using cv::videocapture::set(), in order to run faster on high-resolution cameras. Depending on the camera model, driver, or system, OpenCV might not change the properties of your camera. It is not important for this project, so no need to worry if it does not work with your camera. We can put this code in the main() function of your main_desktop.cpp: int cameranumber = 0; if (argc > 1) cameranumber = atoi(argv[1]); // Get access to the camera. cv::videocapture camera; camera.open(cameranumber); if (!camera.isopened()) { std::cerr << "ERROR: Could not access the camera or video!" << std::endl; exit(1); } // Try to set the camera resolution. camera.set(cv::cv_cap_prop_frame_width, 640); camera.set(cv::cv_cap_prop_frame_height, 480); After the webcam has been initialized, you can grab the current camera image as

8 a cv::mat object (OpenCV's image container). You can grab each camera frame by using the C++ streaming operator from your cv::videocapture object into a cv::mat object, just like if you were getting input from a console. OpenCV makes it very easy to load a video file (AVI or MPEG file) and use it instead of a webcam. The only difference to the code would be that we have to create the cv::videocapture object with the video filename, such as camera.open("my_video.avi"), rather than the camera number, such as camera.open(0). Both methods create a cv::videocapture object that can be used in the same way Summary OpenCV is equipped with lots of algorithms associated with digital image processing as well as computer vision and such algorithmic execution can be explored from the online documentation easily available. OpenCV is having rich documentation in tutorials, user guide, books, quick start, reference and wiki modules. Moreover, a huge community is there to keep track of the bugs and troubleshooting.