Design of automatic cotton picking robot with Machine vision using Image Processing algorithms

摘要

Agricultural sector is desperately in need of engineering outcomes such as Automation, Information Technology and more recently Robotic Technology. Cotton Cultivation in India occupying a big share in commercial crops is facing a major problem of picking the cotton from the plants by the labor as the labor costs are increasing these days. This paper aims at achieving a prominent solution with the use of Machine vision together with Image Processing and Microcontrollers for identification, recognition, and processing of the cotton image as such and picking the cotton with robotic arms to yield maximum production in a day per hectare. Research and development in perceptual system for robots enabled the agricultural sector to catch hold of the technology in reducing the overall cost. These intelligent robots use variety of visual sensors to detect objects with respect to their identity, position, color, orientation in 3D pattern at the fields. This paper also proposes at the new algorithms in Image processing of the cotton to extract the feature, modeling and matching. These are Artificial Intelligence for Robotic Vision, Image Processing for Segmentation, feature measurement such as invariants, size and shape, texture and scene analysis and controlling the robotic arms in desired angle.