A Target Image Recognition Method of Agricultural Harvesting Robot using Template Matching in Low Light Environment | [Pakistan Journal of Agricultural Sciences • 2025]

Author(s):

1. Hongan Chen: Jiangmen Polytechnic,529000, Jiangmen,China

2. Zongfu Zhang: Jiangmen Polytechnic,529000, Jiangmen,China

3. Qingjia Luo: Jiangmen Polytechnic,529000, Jiangmen,China : Faculty of Data Science, City University of Macau, Avenida Padre Tomás Pereira Taipa, Macau, 999078, China

4. Rongbin Chen: Jiangmen Polytechnic,529000, Jiangmen,China

5. Yang Zhao: Guangdong University of Science and Technology,523083, Dongguan,ChinaIntelligent Manufacturing and Environmental Monitoring Engineering Technology Research Center of Dongguan City, Guangdong University of Science and Technology,Dongguan 523000,China

Abstract:

In low-light scenarios, agricultural harvesting robots often encounter significant challenges in visual recognition, leading to inaccurate identification of harvesting targets. To address this issue, this paper proposes a novel target recognition method for agricultural harvesting robots in low-light conditions. Specifically, an eye-to-hand model is developed for the agricultural picking robot to acquire target images. The acquired images are then preprocessed to enhance their quality. To determine the optimal threshold value, the maximum inter-class variance method is employed to analyze the foreground and background of the preprocessed target images. Subsequently, the target images are segmented based on the threshold value, and the edges of the target images are detected. Feature points are extracted from the edges, and the target recognition is completed through template matching. Experimental results demonstrate that the proposed method achieves a maximum recognition accuracy of 98.56% and significantly reduces recognition time, with a maximum recognition time of no more than 0.5 seconds. This method exhibits superior performance in target recognition for agricultural harvesting robots under low-light conditions. The application of this study has the potential to revolutionize agricultural automation by enabling precise and efficient harvesting in low-light environments, thereby increasing crop yields and reducing labor costs. This innovative approach is expected to be widely adopted in the agricultural industry, enhancing overall productivity and sustainability in farming practices.

Page(s): 79-90

DOI: 10.21162/PAKJAS/25.66

Published: Journal: Pakistan Journal of Agricultural Sciences, Volume: 62, Issue: 1, Year: 2025

Keywords:

Edge detection , template matching , threshold segmentation , target image preprocessing , agricultural harvesting robots , Low light scenarios

References:

[1] Arikapudi S.G.,Vougioukas S.G. .2023 .Robotic tree-fruit harvesting with arrays of cartesian arms: a study of fruit pick cycle times. Computers and Electronics in Agriculture, 211 : 108023-108023.

[2] Bah ,M.D. A.,Hafiane R.,Canals R. .2023 .Hierarchical graph representation for unsupervised crop row detection in images. Expert Systems with Applications, 216 : 119478-119478.

[3] Fang C.,Dandan L.,Lin L. .2021 .Target recognition and positioning algorithm of picking robot based on deep learning. Electronic Measurement Technology, 44 : 162-167.

[4] Fang W.,Peng L. .2020 .Simulation of Automatic Recognition of Weak Point Targets in Infrared Images under Complex Background. Computer Simulation, 37 : 471-475.

[5] Guojun C.,Wei C.,Hanqi Y. .2019 .Research on target tracking method of underwater vehicle based on monocular vision and depth learning. Machine Tools and Hydraulics, 47 : 79-82.

[6] Guoliang D.,Lei Z.,Shan X. .2022 .Target recognition and location of steel bar binding robot based on deep learning. Electronic Measurement Technology, 45 : 35-44.

[7] Hualin Y.,Long C.,Miaoting C.,Zhibin M.,Fang D.,Maozhen L.,Xiangrong L. .2019 .Tender Tea Shoots Recognition and Positioning for Picking Robot Using Improved YOLO-V3 Model. IEEE Access, 7 : 180998-180011.

[8] Huan Z.,Wei H. .2020 .Research on target recognition, tracking and positioning of visual robot based on soccer match rules. Journal of Baoshan University, 39 : 50-55.

[9] Haiwen Z.,Feng L.,Yachuan Z.,Xingyue Q. .2019 .Rapid recognition method for loading state of robot elevator hall door based on YOLO model. Packaging Engineering, 40 : 180-185.

[10] Jin W.,Ruirong W.,Xiaohong L. .2021 .Study on target recognition method of tomato picking robot. Jiangsu Agricultural Sciences, 49 : 217-222.

[11] Kai W.,Guoting X.,Liwei L.,Yong L.,Xiao F. .2019 .Application of edge detection and evidence theory in robot target recognition. Laboratory research and exploration, 38 : 29-33.

[12] Cheng F.,Qingchun S.,Yuhuan L.,Yajun R.,Mengfei X.,Lijia X. .2023 .YOLACTFusion: An instance segmentation method for RGB-NIR multimodal image fusion based on an attention mechanism. Computers and Electronics in Agriculture, 213 : 108186-108186.

[13] Jidong L.,Hao X.,Liming X.,Ling Z.,Hailong R.,Biao Y.,N. Y.,Zhenghua M. .2022 .Recognition of fruits and vegetables with similar-color background in natural environment: A survey. Journal of Field Robotics, 6 : 888-904.

[14] Longzhi Z.,Dongmei W. .2020 .A Mobile Robot Recognize Blurred Targets via an Improved Histogram Equalization. IOP Conference Series Materials Science and Engineering, 853 : 012042.

[15] Parvin M.,Jafar A.V.,Keyvan A.V. .2023 .Robotic date fruit harvesting using machine vision and a 5‐DOF manipulator. Journal of Field Robotics, 40 : 1408-1423.

[16] Mejia G.,Andrés M.O.,Flores G. .2023 .Strawberry localization in a ridge planting with an autonomous rover. Engineering Applications of Artificial Intelligence, 119 : 105810-105810.

[17] Pengcheng W.,Bo W. .2020 .Multi-sensor detection and control network technology based on parallel computing model in robot target detection and recognition. Computer Communications, 159 : 215-221.

[18] Pengyu Z.,Zongyan W.,Xiao H.,Peiliao D. .2022 .Research on target recognition and grabber of delta robot based on machine vision. Tool Engineering, 56 : 113-117.

[19] Qiang B.,Jing Y.,Qisong S.,Zhiang L.,Xingxing Z. .2020 .Object detection recognition and robot grasping based on machine learning: A survey. IEEE ACCESS, 8 : 181855-181879.

[20] Weimin. R. ,Baojiang L.,Huanjun L.,Hang O,Yueyu S. .2023 .Ground penetrating radar (GPR) identification method for agricultural soil stratification in a typical mollisols area of northeast china. Chinese Geographical Science, 33 : 664-678.

[21] Guntitat S.,Tanyatep T.,Thanawat L.,Puchong S.,Sarana N.,Poramate M.,D. Nat.. M. .2022 .Visual goal human-robot communication framework with few-shot learning: a case study in robot waiter system. IEEE Transactions on Industrial Informatics, 18 : 1883-1891.

[22] Shengshu L.,Guohua G.,Junzhou W.,Xingjian C. .2019 .An object recognition algorithm based on machine vision of collaborative robot baxter. Electronics Optics & Control, 26 : 105-99.

[23] Shuhua L.,Huixin X.,Qi L.,Fei Z.,Kun H. .2021 .A robot object recognition method based on scene text reading in home environments. Sensors, 21 : 1919-1919.

[24] Taixiong Z.,Mingzhe J.,Mingchi F. .2021 .Vision based target recognition and location for picking robot:A review. Chinese Journal of Scientific Instrument, 49 : 28-51.

[25] Hao W.,Zeming F.,Xiaojun Y.,Pengbo K. Meilin W.,Xilai Z. .2022 .A real-time branch detection and reconstruction mechanism for harvesting robot via convolutional neural network and image segmentation. Computers and Electronics in Agriculture, 192 : 106609-106609.

[26] Yuqing G.,Qingjun Z.,Nan X.,Xiaotian S.,Huiwei X. .2021 .Research on target recognition of autonomous underwater vehicle based on image enhancement. China Measurement & Test, 47 : 47-52.

[27] Zhenyu S.,Xiaoming G.,Xiaoyang Z.,Jiangxue H.,Jian H. .1948 .Research on robot target recognition based on deep learning. Journal of Physics: Conference Series, : 012056-012056.

Citations

Downloads

Views