Green Pepper Fruits Counting: Based On Improved DeepSort and Optimised Yolov5s

Original video: https://youtu.be/WToI9xvkjIE

Green pepper, a significant commercial crop, can undergo multiple harvests based on fruit maturity during the growing season (Stein et al., 2016), and pepper yields exhibit considerable variation across different harvesting periods. Consequently, pre-harvest estimation of green pepper yield can significantly aid in optimising harvest processes, labour management, transportation, and storage conditions (He et al., 2022). Currently, green pepper yield estimation predominantly depends on manual sampling, a time-consuming and labour-intensive method (Aggelopoulou et al., 2010; Wulfsohn et al., 2012).

This video proposes an automatic counting method for green pepper fruits based on object detection and multi-object tracking algorithm. Green pepper fruits have colors similar to leaves and are often occluded by each other, posing challenges for detection. Based on the YOLOv5s, the CS_YOLOv5s model is specifically designed for green pepper fruit detection. In the CS_YOLOv5s model, a Slim-Nick combined with GSConv structure is utilised in the Neck to reduce model parameters while enhancing detection speed. Additionally, the Convolutional Block Attention Module (CBAM) attention mechanism is integrated into the Neck to enhance the feature perception of green peppers at various locations and enhance the feature extraction capabilities of the model.

Keywords: DeepSORT, Deep Learning, Green Pepper, Fruit Counting, Tracking

Citation: Du P, Chen S, Li X, Hu W, Lan N, Lei X and Xiang Y (2024) Green pepper fruits counting based on improved DeepSort and optimized Yolov5s. Front. Plant Sci. 15:1417682. doi: 10.3389/fpls.2024.1417682

Received: 15 April 2024; Accepted: 31 May 2024;

Published: 16 July 2024.

Attribution 4.0 International — CC BY 4.0 - Creative Commons