Early bruising detection of 'Korla' pears by low-cost visible-LED structured-illumination reflectance imaging and feature-based classification models

Mengwen Mei; Zhonglei Cai; Xinran Zhang; Chanjun Sun; Junyi Zhang; Huijie Peng; Jiangbo Li; Ruiyao Shi; Wei Zhang

doi:10.3389/fpls.2023.1324152

Early bruising detection of 'Korla' pears by low-cost visible-LED structured-illumination reflectance imaging and feature-based classification models

Front Plant Sci. 2023 Nov 16:14:1324152. doi: 10.3389/fpls.2023.1324152. eCollection 2023.

Authors

Mengwen Mei¹, Zhonglei Cai¹, Xinran Zhang¹, Chanjun Sun², Junyi Zhang¹, Huijie Peng^{1

3

4}, Jiangbo Li^{5

6}, Ruiyao Shi⁵, Wei Zhang⁷

Affiliations

¹ College of Mechanical and Electrical Engineering, Shihezi University, Shihezi, China.
² Jiangsu Province and Education Ministry Co-sponsored Synergistic Innovation Center of Modern Agricultural Equipment, Jiangsu University, Zhenjiang, China.
³ Xinjiang Production and Construction Corps Key Laboratory of Modern Agricultural Machinery, Shihezi, China.
⁴ Engineering Research Center for Production Mechanization of OasisCharacteristic Cash Crop, Ministry of Education, Shihezi, China.
⁵ Intelligent Equipment Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing, China.
⁶ National Engineering Research Center for Information Technology in Agriculture, Beijing, China.
⁷ Department of Computer Technology and Science, Anhui University of Finance and Economics, Bengbu, China.

Abstract

Introduction: Nondestructive detection of thin-skinned fruit bruising is one of the main challenges in the automated grading of post-harvest fruit. The structured-illumination reflectance imaging (SIRI) is an emerging optical technique with the potential for detection of bruises.

Methods: This study presented the pioneering application of low-cost visible-LED SIRI for detecting early subcutaneous bruises in 'Korla' pears. Three types of bruising degrees (mild, moderate and severe) and ten sets of spatial frequencies (50, 100, 150, 200, 250, 300, 350, 400, 450 and 500 cycles m^-1) were analyzed. By evaluation of contrast index (CI) values, 150 cycles m^-1 was determined as the optimal spatial frequency. The sinusoidal pattern images were demodulated to get the DC, AC, and RT images without any stripe information. Based on AC and RT images, texture features were extracted and the LS-SVM, PLS-DA and KNN classification models combined the optimized features were developed for the detection of 'Korla' pears with varying degrees of bruising.

Results and discussion: It was found that RT images consistently outperformed AC images regardless of type of model, and LS-SVM model exhibited the highest detection accuracy and stability. Across mild, moderate, severe and mixed bruises, the LS-SVM model with RT images achieved classification accuracies of 98.6%, 98.9%, 98.5%, and 98.8%, respectively. This study showed that visible-LED SIRI technique could effectively detect early bruising of 'Korla' pears, providing a valuable reference for using low-cost visible LED SIRI to detect fruit damage.

Keywords: classification; early bruise detection; machine learning; pears; visible LED structured illumination.

Grants and funding

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. Authors are grateful for the National Natural Science Foundation of China (31972152), Jiangsu Province and Education Ministry Co-sponsored Synergistic Innovation Center of Modern Agricultural Equipment (XTCX2001) and the Bingtuan Science and Technology Program (2022DB004) for providing funding for the study.