1
|
Zhang Z, Lei X, Huang K, Sun Y, Zeng J, Xyu T, Yuan Q, Qi Y, Herbst A, Lyu X. Multi-scenario pear tree inflorescence detection based on improved YOLOv7 object detection algorithm. Front Plant Sci 2024; 14:1330141. [PMID: 38317836 PMCID: PMC10840500 DOI: 10.3389/fpls.2023.1330141] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 10/30/2023] [Accepted: 12/27/2023] [Indexed: 02/07/2024]
Abstract
Efficient and precise thinning during the orchard blossom period is a crucial factor in enhancing both fruit yield and quality. The accurate recognition of inflorescence is the cornerstone of intelligent blossom equipment. To advance the process of intelligent blossom thinning, this paper addresses the issue of suboptimal performance of current inflorescence recognition algorithms in detecting dense inflorescence at a long distance. It introduces an inflorescence recognition algorithm, YOLOv7-E, based on the YOLOv7 neural network model. YOLOv7 incorporates an efficient multi-scale attention mechanism (EMA) to enable cross-channel feature interaction through parallel processing strategies, thereby maximizing the retention of pixel-level features and positional information on the feature maps. Additionally, the SPPCSPC module is optimized to preserve target area features as much as possible under different receptive fields, and the Soft-NMS algorithm is employed to reduce the likelihood of missing detections in overlapping regions. The model is trained on a diverse dataset collected from real-world field settings. Upon validation, the improved YOLOv7-E object detection algorithm achieves an average precision and recall of 91.4% and 89.8%, respectively, in inflorescence detection under various time periods, distances, and weather conditions. The detection time for a single image is 80.9 ms, and the model size is 37.6 Mb. In comparison to the original YOLOv7 algorithm, it boasts a 4.9% increase in detection accuracy and a 5.3% improvement in recall rate, with a mere 1.8% increase in model parameters. The YOLOv7-E object detection algorithm presented in this study enables precise inflorescence detection and localization across an entire tree at varying distances, offering robust technical support for differentiated and precise blossom thinning operations by thinning machinery in the future.
Collapse
Affiliation(s)
- Zhen Zhang
- School of Agricultural Engineering, Jiangsu University, Zhenjiang, China
- Institute of Agricultural Facilities and Equipment, Jiangsu Academy of Agricultural Sciences, Nanjing, China
- Key Laboratory of Modern Horticultural Equipment, Ministry of Agriculture and Rural Affairs, Nanjing, China
| | - Xiaohui Lei
- Institute of Agricultural Facilities and Equipment, Jiangsu Academy of Agricultural Sciences, Nanjing, China
- Key Laboratory of Modern Horticultural Equipment, Ministry of Agriculture and Rural Affairs, Nanjing, China
| | - Kai Huang
- Institute of Agricultural Facilities and Equipment, Jiangsu Academy of Agricultural Sciences, Nanjing, China
- Key Laboratory of Modern Horticultural Equipment, Ministry of Agriculture and Rural Affairs, Nanjing, China
| | - Yuanhao Sun
- Institute of Agricultural Facilities and Equipment, Jiangsu Academy of Agricultural Sciences, Nanjing, China
- Key Laboratory of Modern Horticultural Equipment, Ministry of Agriculture and Rural Affairs, Nanjing, China
| | - Jin Zeng
- Institute of Agricultural Facilities and Equipment, Jiangsu Academy of Agricultural Sciences, Nanjing, China
- Key Laboratory of Modern Horticultural Equipment, Ministry of Agriculture and Rural Affairs, Nanjing, China
| | - Tao Xyu
- Institute of Agricultural Facilities and Equipment, Jiangsu Academy of Agricultural Sciences, Nanjing, China
- Key Laboratory of Modern Horticultural Equipment, Ministry of Agriculture and Rural Affairs, Nanjing, China
| | - Quanchun Yuan
- Institute of Agricultural Facilities and Equipment, Jiangsu Academy of Agricultural Sciences, Nanjing, China
- Key Laboratory of Modern Horticultural Equipment, Ministry of Agriculture and Rural Affairs, Nanjing, China
| | - Yannan Qi
- Institute of Agricultural Facilities and Equipment, Jiangsu Academy of Agricultural Sciences, Nanjing, China
- Key Laboratory of Modern Horticultural Equipment, Ministry of Agriculture and Rural Affairs, Nanjing, China
| | - Andreas Herbst
- Institute for Chemical Application Technology of JKI, Braunschweig, Germany
| | - Xiaolan Lyu
- Institute of Agricultural Facilities and Equipment, Jiangsu Academy of Agricultural Sciences, Nanjing, China
- Key Laboratory of Modern Horticultural Equipment, Ministry of Agriculture and Rural Affairs, Nanjing, China
| |
Collapse
|
2
|
Qin F, Bu X, Liu Y, Liang X, Xin J. Foreign Object Debris Automatic Target Detection for Millimeter-Wave Surveillance Radar. Sensors (Basel) 2021; 21:s21113853. [PMID: 34199670 PMCID: PMC8199731 DOI: 10.3390/s21113853] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 05/06/2021] [Revised: 05/20/2021] [Accepted: 05/31/2021] [Indexed: 11/16/2022]
Abstract
Foreign Object Debris (FOD) refers to any foreign material on the airfield that may injure and threaten the aircraft and airport system. Due to the complex background on the airfield pavement and weak target echoes in long-distance monitoring, it is not easy to detect objects of various types and sizes. The existing FOD radar system's detection method has a short effective range, and the detectable objects' radar cross-section intensity is no less than -20 dBsm. In this paper, we propose an integrated FOD automatic target detection algorithm for millimeter-wave (MMW) surveillance radar to improve small target detection under long-range conditions of over 660 m. The signal form of FOD and a clutter model of ground clutter received by millimeter-wave radar are primarily utilized and established theoretically. The runway edge detection means that it is employed based on the in-continuity features as the runway region of interest during the automatic extraction step. Following the clutter map constant false alarm detection algorithm, we utilize a time-domain algorithm that functions as the vital detection processor. Moreover, an explicit definition of the FOD detection performance is developed in a characteristic quantitative way. This criterion involves an absolute reference value for all FOD radar systems. The well-designed FOD frequency-modulated continuous-wave MMW surveillance radar is utilized, and actual experiments are carried out in a real airport in Beijing, China. The results validate the proposed method's effectiveness and the superior performance of FOD target detection in long-range situations.
Collapse
Affiliation(s)
- Fei Qin
- National Key Lab of Microwave Imaging Technology, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, China; (F.Q.); (X.B.); (Y.L.); (J.X.)
- School of Electronics, Electrical and Communication Engineering, University of Chinese Academy of Sciences, Beijing 100049, China
| | - Xiangxi Bu
- National Key Lab of Microwave Imaging Technology, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, China; (F.Q.); (X.B.); (Y.L.); (J.X.)
| | - Yunlong Liu
- National Key Lab of Microwave Imaging Technology, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, China; (F.Q.); (X.B.); (Y.L.); (J.X.)
| | - Xingdong Liang
- National Key Lab of Microwave Imaging Technology, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, China; (F.Q.); (X.B.); (Y.L.); (J.X.)
- School of Electronics, Electrical and Communication Engineering, University of Chinese Academy of Sciences, Beijing 100049, China
- Correspondence:
| | - Jihao Xin
- National Key Lab of Microwave Imaging Technology, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, China; (F.Q.); (X.B.); (Y.L.); (J.X.)
- School of Electronics, Electrical and Communication Engineering, University of Chinese Academy of Sciences, Beijing 100049, China
| |
Collapse
|