Chili Plant Classification Using Transfer Learning Models Through Object Detection: Lecture Notes in Electrical Engineering

This study presents the use of a Convolutional Neural Network (CNN) based detector to detect chili and its leaves in the chili plant image. Detecting chili on its plant is essential for the development of robotic vision and monitoring. Thus, helps us supervise the plant growth, furthermore, analyses their productivity and quality. This paper aims to develop a system that can monitor and identify bird’s eye chili plants by implementing machine learning. First, the development of methodology for efficient detection of bird’s eye chili and its leaf was made. The image labeling will provide the source of images between chili and leaf. The dataset would be split into training, verification and test set for 70:20:10%, correspondingly. The images YOLO V4 Darknet was implemented to train the dataset. After a series of experiments were conducted, the model is compared with other transfer learning models like YOLO V4 Tiny, Faster R-CNN, and EfficientDet. The classification performance of these transfer learning models has been calculated and compared with each other. The experimental result would discuss on part of hyper parameter optimization and transfer learning application. Firstly, the optimization of hyper parameter shows that the YOLO V4 Darknet model achieves mAP of 76.54%, followed by EfficientDet at 73.66% for 512 × 512 input layers. Next, the application of transfer learning. The result shows that YOLO V4 Darknet achieves highest mAP value, 75.69% follow by EfficientDet, with mAP of 71.85%.