Real-time fruit detection using deep neural networks on CPU (RTFD): An edge AI application
文献类型: 外文期刊
第一作者: Mao, DianHui
作者: Mao, DianHui;Sun, Hao;Li, XueBo;Yu, XuDong;Mao, DianHui;Zhang, QingChuan;Wu, JianWei
作者机构:
关键词: PicoDet-S; Lightweight; Edge devices; Quantization; Android
期刊名称:COMPUTERS AND ELECTRONICS IN AGRICULTURE ( 影响因子:6.757; 五年影响因子:6.817 )
ISSN: 0168-1699
年卷期: 2023 年 204 卷
页码:
收录情况: SCI
摘要: Intelligent harvesting can greatly reduce the manual labor for fruit harvesting and is one of the main application areas of intelligent agricultural machinery. Smart edge sensors must detect the growth status of tomatoes and strawberries in real-time for intelligent harvesting to be possible. Due to edge devices' relatively low image processing power, it remains challenging to implement deep learning algorithms for real-time detection in field applications. Therefore, scholars are strongly interested in utilizing computationally small, accurate, and quick models for edge harvesting machinery. In this research, we propose RTFD, a lightweight algorithm for edge CPU devices that detects fruit. Based on the PicoDet-S model, RTFD optimizes the efficiency of real-time detection for edge CPU computing devices by enhancing the model's structure, loss function and activation function. The experimental results demonstrate that the mAP of RTFD for tomato and strawberry datasets at various growth phases is 0.78 and 0.96, respectively, with a computational effort of 1.44 GFLOPs. RTFD improves mAP by 1.9% and 2.3% on the tomato and strawberry datasets, respectively, with minimal loss in computation and model parameters, making it more suitable for edge device deployment. Meanwhile, we proposed an approach for 8-bit quantization-aware training of the Conv2D and Linear layers of the model for improved edge deployment. This approach quantizes 32-bit floating-point values in the Conv2D and Linear layers into 8-bit integers, thereby reducing the model's size and accelerating detection speed. The testing findings indicate that the quantized RTFD model has a size of 1.33 MB and a CPU detection speed of around 11 FPS. Using the quantified trained model, we developed the Android application RTFD-CPU to evaluate the model's performance in test scenarios. RTFD-CPU was assessed on the smartphone Redmi K30pro (Qualcomm Snapdragon 865 CPU and 8 GB of RAM) with a detection speed of 19 FPS utilizing only the CPU. In summary, the proposed RTFD has great potential for intelligent picking machines, and the concept of redesigning the model structure, loss function, and activation function, as well as training by quantization to speed up the detection of deep neural networks, is anticipated to be successfully implemented in edge computing.
分类号:
- 相关文献
作者其他论文 更多>>
-
A hyperspectral band selection method based on sparse band attention network for maize seed variety identification
作者:Zhang, Liu;Wei, Yaoguang;Liu, Jincun;An, Dong;Zhang, Liu;Wei, Yaoguang;Liu, Jincun;An, Dong;Zhang, Liu;Wei, Yaoguang;Liu, Jincun;An, Dong;Zhang, Liu;Wei, Yaoguang;Liu, Jincun;An, Dong;Wu, Jianwei;Wu, Jianwei
关键词:Hyperspectral imaging; Band selection; Attention mechanism; Deep learning; Seed variety identification
-
Maize seed variety identification using hyperspectral imaging and self-supervised learning: A two-stage training approach without spectral preprocessing
作者:Zhang, Liu;Zhang, Shubin;Liu, Jincun;Wei, Yaoguang;An, Dong;Zhang, Liu;Zhang, Shubin;Liu, Jincun;Wei, Yaoguang;An, Dong;Zhang, Liu;Zhang, Shubin;Liu, Jincun;Wei, Yaoguang;An, Dong;Zhang, Liu;Zhang, Shubin;Liu, Jincun;Wei, Yaoguang;An, Dong;Wu, Jianwei;Wu, Jianwei
关键词:Seed classification; Hyperspectral imaging; Self-supervised learning; Deep learning; Spectral analysis
-
Development of an evapotranspiration estimation method for lettuce via mobile phones using machine vision: Proof of concept
作者:Zhangzhong, Lili;Gao, Hairong;Zheng, Wengang;Li, Jingjing;Wang, Dequn;Zhangzhong, Lili;Gao, Hairong;Zheng, Wengang;Li, Jingjing;Wang, Dequn;Wu, Jianwei
关键词:Image processing; Machine learning; Crop coefficient; Crop evapotranspiration
-
Acid-promoted lignin reductive depolymerization under mild conditions via a condensation minimizing approach: From organosolv lignin to woody biomass
作者:Wei, Ni;Qi, Shuang;Wang, Guanhua;Ge, Jing;Sun, Hao;Si, Chuanling;Sui, Wenjie;Parvez, Ashak Mahmud;Jia, Hongyu;Si, Chuanling;Wang, Guanhua;Si, Chuanling
关键词:Lignin reductive depolymerization; Liquid fuels; Condensation minimizing; Oligomer characterization; Lignin -first reductive fractionation
-
Effect and Mechanism of Micro-Nano Aeration Treatment on a Drip Irrigation Emitter Based on Groundwater
作者:Li, Rui;Han, Qibiao;Hu, Yawei;Li, Rui;Han, Qibiao;Dong, Conghui;Li, Hao;Sun, Hao;Li, Hui;Li, Peng;Nan, Xi;Hu, Yawei
关键词:drip irrigation; emitter clogging; blockage; composition; average flow ratio; micro-nano bubbles
-
Whole genome sequencing of Aoluguya reindeer (Rangifer tarandus) in China
作者:Shi, Lulu;Shi, Zheng;Hu, Mingyue;Wu, Mao;Sun, Hao;Yan, Shouqing;Quan, Xinjiao;Tian, Laiming;Qin, Lihong;Zhao, Zhongli
关键词:Aoluguya reindeer; whole genome sequencing; SNP; genetic relationship; genetic resource
-
CFD SIMULATION OF CIRCULATING-AIRFLOW DISTRIBUTION INSIDE CANOPY FROM NOVEL AIR-ASSISTED SPRAYER IN ORCHARD
作者:Sun, Hao;Zheng, He;Yu, Hongfeng;Qiu, Wei;Cao, Yubin;Zhang, Zhengwei;Lv, Xiaolan
关键词:Circulating air-assisted spraying; Fruit-tree canopy; Inlet and outlet airflow parameters; Simulation