This thesis presents the development of a machine vision system aimed at automating the Pick and Place operation in small-scale industries, thereby reducing reliance on human workforce. A key objective of this research was to improve the ergonomic conditions within such industries. The vision system's development involved two distinct approaches: Image Processing and Deep Learning techniques, each offering unique advantages and disadvantages. While the Image Processing techniques exhibited robustness, they faced limitations in accurately detecting objects in a manufacturing process. Consequently, Deep Learning, specifically the You Only Look Once (YOLO) algorithm, was adopted for object detection due to its notable benefits, including high speed, accuracy, real-time capabilities, and ease of deployment. The research was conducted utilizing various software tools, such as MATLAB, Python, and Google Collab. The outcomes of this thesis encompass the successful creation of the machine vision system, capable of accurately detecting objects and providing their respective coordinates. Additionally, emphasis was placed on the system's ability to correctly classify objects and assign higher confidence scores, indicative of its proficiency in identifying objects with enhanced accuracy. By implanting this system, the research highlights its potential for enhancing the ergonomics of small-scale industries, contributing to the overall efficiency and productivity of their Pick and Place operation.