Abstract

In video surveillance, detection of moving object is crucial and basic task as its output is an input to the next-level such as tracking and behavior analysis. Detecting camouflage object in a video is the biggest challenge. Computer vision methods are already largely used. With recent development in deep Learning, development of object detection application has become easier. TensorFlow’s Object Detection API is an open source model that is developed and pretrained on the datasets. Thus, in this article for detecting camouflage object under different environment tensor flow is used. Quantitative results were obtained on standard datasets and compared with the other algorithms. TensorFlow gives excellent results as compared to other algorithms.