Apple packing process still is done by hand. Along the apple conveyers, workers are placed to monitor the passing apples and remove the rotted, damaged, diseased apples and foreign materials. This labor intensive work in a wet and dirty environment is not only tedious and causes fatigue to eyes, but also difficult to prevent the undesirable objects from passing through the line. It has been an increasing issue and demand in the fruit packing industry to solve the fruit quality and labor problems by using advanced technology. Technology of machine vision in conjunction with robotics provides a great potential to assist humans to do the fruit inspection work. By using video cameras and image processing systems to identify undesirable objects and then guiding robotic arms to pick out these objects is a possible and practical solution to the problem. The objective of the research is to develop effective vision methods to extract features and recognize the undesirable objects on the apple process lines, and determine the locations for robotics actions.
Anticipated Results:The anticipated result will be that the camera can recognize the major types of undesirable objects among the normal apples in the viewing plane through digital image processing, and locate the objects for a robotics system. The success of this research will not only be a practical high-tech application to solve real problems, but also be a great step in technological innovation in the fruit industry. It will have a direct positive impact on packing line efficiency, healthy work environment, fruit quality, and the competition power of apples in the world market.