The requirements for more efficient and easily modifiable techniques have provoked the fast advancement in the domain of robotics. The improvement of canny robots prompts the capacity of them to turn into an administrator profoundly proficient and ready to adjust to a wide scope of issues. In any case, notwithstanding the few automated arrangements accessible, most of the current modern robots do not utilize the Robotic Operative System (ROS) and have restrictions as far as self-sufficiently right mistakes during their assignments. Controlling a robotic arm for applications such as object segmentation with the utilization of vision sensors would require vigorous picture processing and calculation to perceive and distinguish the object when using an image processing heavy approach, while a more traditional approach relies on sensors and partial automation in most cases. This paper is coordinated toward compiling the important computer vision techniques for pick and place operation along with the underlying factors that make them better than traditional techniques.