Computer vision is one of the recent approaches used to increase robotic efficiency. Robotics has proved its effectiveness in working in radiation and nuclear environments in order to increase human safety. Nowadays, robots have become cheaper whereas the (re)programming and (re)training cost may exceed the robot installation cost. The current research investigates an application of perceptive Vision-based offline programming assisted by visual cues to control a robotic arm in a hazardous and radioactive environment to ease the robot programming and fill the gap of operator absence in the workspace. An experiment has been conducted to pick an object and place in a safe area. The position and orientation have been estimated without deep interference of the operator in mathematical calculation, visual feedback is provided to the operator, forward and inverse kinematics are calculated. A pick and place experiment has been conducted using MATLAB/SIMULINK program and embedded microcontroller system conducted to show the validity of the proposed system and prove its success.