Artificial Intelligence-Based Detection System For Fruit Identification

The process of identifying fruit involves identifying and categorizing the final product only based on its physical characteristics, such as color, shape, size, texture, and sample. The use of the fruit identification era in monitoring contexts would possibly enhance ethical concerns. The system may struggle with diverse fruit appearances, varying lighting conditions, and overlapping or damaged fruit, potentially leading to decreased accuracy and challenges in real-international deployment and scalability. The novel Pine cone optimization driven Redefined CNN (PCO-RCNN) was suggested as a detection method for fruit identification to overcome these limitations. The fruit image data were collected in FIDS30 and pre-processed it with a Histogram of Oriented Gradients (HOG). The experiment is done using the Python platform and comparison of existing methods, our suggested method achieved high accuracy with 95%. This study promotes the usage of local and unusual culmination by ensuring correct identification.