| Popis: |
Object detection is one of the most important research areas in image processing. Here, we propose to employ dual camera system to precisely find the distance, the height, the width of object that are applicable in areas like autonomous system navigation, vehicle navigation, etc. One of the most popular approach is the passive approach; however, its performance is usually limited by complicated scenes and texture-fewer areas. Therefore, actual and resourceful estimation algorithms of stereo depth have been a long-time problem. To solve those problems, an enhanced methodology that embraces three steps are proposed. Foremost, the images are segmented using thresholding technique, and a blob detection technique is used to find the objects in the scene. Then, the center of gravity is found for the objects. It helps to avoid the rectification process. Finally, the distance of the objects is found using the disparity values of the objects. Experiment results are promising. |
