A Review on Techniques for Multiple Object Recognition from a Scene

Object recognition is a computer vision technique for identifying objects in an image or video. Object recognition is a key output of deep learning and machine learning algorithm. Humans can recognize any object in the real world easily without any effort and hence are able to do their work efficiently. Similarly if robots too become efficient in understanding their environment it will be able to perform many complex tasks better. Some of those tasks include object tracking, industrial inspection, bio imaging, robotic vision etc. One of the trending applications of them all is in the intelligent vehicle system. Object recognition helps in recognizing objects or obstacles in the path of a robot, sign boards and people in case of a driverless car, abnormalities during any kind of inspection etc. Histograms of Oriented Gradients (HOG), CNN, RCNN, Fast R-CNN, Faster R-CNN, FCN, Mask R-CNN, YOLO, and YOLO9000 are some of the techniques discussed here.