Object detection and tracking could be a immense, vivacious however inconclusive and trending area of computer vision. Due to its immense use in official surveillances, tracking modules applied in security and lots of others applications have made researchers to devise a lot of optimized and specialized methods. However, problems are faced in implementing object detection and tracking in real-time; like tracking in real time and giving appropriate optimized results, over dynamic computation to find the efficient performance with respect to time factor, or multiple objects tracking create this task more difficult. Though, several techniques are devised but still lies a lot of scope of improvement, however during this literature review we've seen some illustrious and multiple ways of object detection and tracking. In this method we will be using Tensor Flow and Open CV library and CNN algorithm will be used and we will be labelling the detected layers with accuracy being checked at the same time. For validation purpose live input video will be taken for the same where objects will be getting detected and it can be simulated same for real-time through external hardware added. In the end we see the proper optimized and efficient algorithm for object tracking and detection.
CITATION STYLE
Sujeetha, R., & Mishra, V. (2019). Object detection and tracking using tensor flow. International Journal of Recent Technology and Engineering, 8(1), 3397–3401.
Mendeley helps you to discover research relevant for your work.