Detection of Partially Occluded Area in Images Using Image Segmentation Technique

Computer vision is subfield of artificial intelligence. Detection of occluded area in images is one of the challenging tasks in the computer vision. In this paper, we concentrate on occluded area detection in images using image segmentation techniques. Occlusion is nothing but one object in image is hidden by another object. Detection of occluded area in images using image segmentation techniques. Image segmentation is the process of dividing the images into different regions based on the characteristics of the pixels in the original image and reduces the complexity of analysis. When determining the number of objects in a scene, instance segmentation is an excellent choice. But in semantic segmentation algorithms if all that is required is to group items that belong to the same class. Image segmentation has many applications in a number of fields, including gaming, robotics, autonomous vehicles, agriculture, object detection, and pedestrian detection. In this research work, we concentrate on detection of occluded area in images using Mask R-CNN. The experiment will show better segmentation results corresponding to the manually labeled occluded area.