Automation in fruit ripening through Image Processing

One popular fruit grown and consumed worldwide is the mango. Currently, mangoes are manually sorted based on their ripeness level for commercial purposes, leading to inaccuracies and increased labor costs in the processing sector. An alternative method proposed involves using visual characteristics to categorize mangoes, with Badami mangoes used as a case study. This approach can be applied to any variety of mango that undergoes color changes during ripening. By analyzing the average gray-scale values of pixels in images and developing algorithms using Python programming with Open-Source Computer Vision Library, it was found that there is a significant shift in gray-scale values during mangos' ripening process. When compared with manual results, this technique demonstrated an accuracy rate of 90.90%, assuming ambient lighting conditions when capturing images.