Design of Citrus Fruit Detection System Based on Mobile Platform and Edge Computer Device

Abstract

Citrus fruit detection can provide technical support for fine management and yield determination of citrus orchards. Accurate detection of citrus fruits in mountain orchards is challenging because of leaf occlusion and citrus fruit mutual occlusion of different fruits. This paper presents a citrus detection task that combines UAV data collection, AI embedded device, and target detection algorithm. The system used a small unmanned aerial vehicle equipped with a camera to take full-scale pictures of citrus trees; at the same time, we extended the state-of-the-art model target detection algorithm, added the attention mechanism and adaptive fusion feature method, improved the model’s performance; to facilitate the deployment of the model, we used the pruning method to reduce the amount of model calculation and parameters. The improved target detection algorithm is ported to the edge computing end to detect the data collected by the unmanned aerial vehicle. The experiment was performed on the self-made citrus dataset, the detection accuracy was 93.32%, and the processing speed at the edge computing device was 180 ms/frame. This method is suitable for citrus detection tasks in the mountainous orchard environment, and it can help fruit growers to estimate their yield.

Strategies for improving the detection accuracy of computerized machine vision considering spatial applications

“…Huang, H et al utilized a combination of computer vision, machine learning, and edge computing to provide an efficient and accurate solution for the citrus detection task. To facilitate the deployment of the model, a pruning approach was used to reduce the computational effort and parameters of the model [16].…”

Application of Convolutional Neural Network-Based Detection Methods in Fresh Fruit Production: A Comprehensive Review

“… Apolo-Apolo et al (2020) used a UAV to monitor citrus in orchards (shown in Figure 12 ) and adopted Faster-R-CNN to develop a system that can automatically detect and estimate the size of citrus fruits and estimate the total yield of citrus orchards according to detection results. To solve the problem of inconvenient data capture in mountain orchards, Huang et al (2022) designed a real-time citrus detection system for yield estimation based on a UAV and the YOLOv5 model. Kalantar et al (2020) presented a system for detection and yield estimation of melons with a UAV.…”

Dragon fruit detection in natural orchard environment by integrating lightweight network and attention mechanism

“…Giang et al (2022) rapidly detected tomatoes based on semantic segmentation neural network of RGB-D image, and the detection accuracy rate was 80.2%. Huang et al (2022) applied the YOLOv5 algorithm to detect the citrus data set collected by UAV, and the detection accuracy rate was 93.32%. Yan et al (2021) proposed a lightweight apple object detection method using improved YOLOv5s to identify grasping and ungrasping apples in apple tree images automatically, and the recognition recall rate, accuracy, AP and F 1 were 91.48%, 83.83%, 86.75% and 87.49%, respectively.…”

“…(2022) rapidly detected tomatoes based on semantic segmentation neural network of RGB-D image, and the detection accuracy rate was 80.2%. Huang et al. (2022) applied the YOLOv5 algorithm to detect the citrus data set collected by UAV, and the detection accuracy rate was 93.32%.…”