•  
  •  
 

Turkish Journal of Agriculture and Forestry

Author ORCID Identifier

PENG ZHOU: 0009-0001-9182-8086

HUAN WNAG', 'WANG: 0009-0001-3807-1205

PENGWEI ZHANG: 0009-0008-7884-8948

SHENGZHI LIU: 0009-0000-2328-1114

WEI CEHN: 0009-0007-5329-2886

WEIHANG ZHAO: 0009-0009-3780-9149

BINGYU CAO: 0009-0009-8892-5385

DOI

10.55730/1300-011X.3197

Abstract

With the continuous development of society, agricultural production technology is also improving. Due to the continuous improvement of economic level, the quality of life has also put forward higher requirements, the quality of agricultural products has gradually received more people’s attention. In the picking and sorting process of traditional fragrant pear, the quality control process is not perfect, and the identification and extraction method is not mature enough. In the complex environment of insufficient light or shade of branches and leaves, the pear picking work cannot quickly and accurately locate the target fruit. Therefore, the traditional identification and extraction methods of fragrant pear were deeply explored in this paper. This paper analyzed its work flow and implementation methods, summarized the shortcomings of traditional recognition and extraction methods based on real activities, and put forward suggestions for improvement. In this paper, the orchard was divided into regions according to environmental factors; all kinds of information of fragrant pear growth were collected, and the information data was put into the measurement node for comprehensive calculation. It combined the Internet of Things and image detection technology to obtain the image data of the fragrant pear growing area in the orchard, and used the visual background extractor (Vibe) algorithm to preprocess, denoise and extract features of the fragrant pear image. In order to verify the practical effect of the optimized fragrant pear recognition and extraction method based on the Internet of Things and image detection technology in this paper, a comparative experiment was conducted. The results show that compared with the traditional pear recognition extraction method, the optimized pear recognition extraction method had better prediction effect in the process of pear image processing, and the recall rate and accuracy rate were increased by about 8% and 12.7%, respectively.

Keywords

Big Data, Fragrant Pear, Internet of Things, Recognition Extraction, Unmanned Aerial Vehicle

First Page

502

Last Page

518

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Download

Share

COinS