Objective Tool for Chili Grading Using Convolutional Neural Network and Color Analysis

Crop Detection and Maturity Classification Using a YOLOv5-Based Image Analysis

“… In the state of the art, three studies offer points of comparison with the approach proposed in this work, focusing on detection and classification [50,52,54], as presented in Table 7. The works by Yin et al [50] and Abubeker et al [52] utilize images where the chili peppers are unobstructed by leaves or branches.…”

“… In the state of the art, three studies offer points of comparison with the approach proposed in this work, focusing on detection and classification [50,52,54], as presented in Table 7. The works by Yin et al [50] and Abubeker et al [52] utilize images where the chili peppers are unobstructed by leaves or branches. In Abubeker et al [52], their model operates within an automated machine tasked with harvesting chilies, resulting in peppers positioned outside the plant, facilitating easier identification.…”

“…The works by Yin et al [50] and Abubeker et al [52] utilize images where the chili peppers are unobstructed by leaves or branches. In Abubeker et al [52], their model operates within an automated machine tasked with harvesting chilies, resulting in peppers positioned outside the plant, facilitating easier identification. Conversely, Zainudin et al [54] developed a dataset within a controlled environment using an artificial plant, with images captured against a white background.…”

“…The dataset for this research consists of images of bird-eye chilies in different positions and backgrounds. The chilies were then picked up by a robot manipulator and sorted by ripeness [52]. The system achieved a mAP of 0.94 and an average accuracy of 0.90 across a total of 1558 images.…”