Abstract
［Objective］ Dry matter and sugar content are two important indicators affecting the quality of kiwifruit.To achieve rapid and accurate detection of these indicators,a non -destructive detection method for key internal quality indicators of kiwifruit is proposed,integrating improved deep convolutional neural network with spectral technology.［Methods］ A spectrometer was used to collect spectral data of kiwifruit,and the data were transformed into two types of two -dimensional images using Gramian Angular Field (GAF ) transformation.An improved convolutional neural network model with multi -dilated convolutions was constructed to predict and analyze key quality indicators of kiwifruit.The model consists of two independent CNN modules connected in parallel to process the two types of two-dimensional images.Multi -dilated convolution strategies,clustering pruning methods,and channel attention mechanisms were incorporated to enhance the model's detection and analysis performance.［Results］］ Compared with other models,the proposed method reduced the average root mean square errors of dry matter and sugar content by 20.59% and 13.04%,respectively,increased the average determination coefficients by 6.45% and 4.34%,respectively,and improved the average relative prediction deviations by 6.99% and 12.78%,respectively.［Conclusion］ The proposed method demonstrates good capability in detecting and analyzing key internal quality indicators of kiwifruit,and provides a valuable reference for non -destructive internal quality testing of kiwifruit.引用格式：陈玮，费文源，栗超，等.融合改进深度卷积神经网络和光谱技术的猕猴桃内部品质无损检测 ［J］.食品与机械，2025，41（6）：136-143.C itation:CHEN Wei,FEI Wenyuan,LI Chao,et al.Non-destructive detection of kiwifruit internal quality based on improved deep convolutional neural network and spectral technology [J].Food & Machinery,2025,41(6):136-143.|
Publication Date
7-3-2025
First Page
136
Last Page
143
DOI
10.13652/j.spjx.1003.5788.2025.60040
Recommended Citation
Wei, CHEN; Wenyuan, FEI; Chao, LI; and Chenxi, WEI
(2025)
"Non-destructive detection of kiwifruit internal quality based on improved deep convolutional neural network and spectral technology,"
Food and Machinery: Vol. 41:
Iss.
6, Article 19.
DOI: 10.13652/j.spjx.1003.5788.2025.60040
Available at:
https://www.ifoodmm.cn/journal/vol41/iss6/19
References
[1] 李玉阔,林苗苗,宋哲,等.中国不同地区中猕 2 号猕猴桃果实品质评价体系的建立 [J].果树学报,2024,41(7):1 368-1 377.LI Y K,LIN M M,SONG Z,et al.Establishment of comprehensive evaluation system for fruit quality of Zhongmi No.2 kiwifruit from different regions of China [J].Journal of Fruit Science,2024,41(7):1 368-1 377.
[2] 张良英,刘林,于立杰,等.软枣猕猴桃果实发育期品质指标及抗氧化能力的变化 [J].食品与机械,2023,39(7):157-164.ZHANG L Y,LIU L,YU L J,et al.Changes of quality characteristics and antioxidant capacities during fruit development of hardy kiwifruit [J].Food & Machinery,2023,39(7):157-164.
[3] 李可,廖茂雯,林籽汐,等.基于多元统计法的不同质量猕猴桃果实品质分析 [J].食品工业科技,2023,44(21):285-292.LI K,LIAO M W,LIN Z X,et al.Analysis of quality in different weight grading kiwifruits based on multivariate statistical method [J].Science and Technology of Food Industry,2023,44(21):285-292.
[4] 万畅.基于高光谱技术猕猴桃内部品质的无损检测 [D].西安:西北农林科技大学,2024:18-20.WAN C.Non-destructive testing of internal quality of kiwifruit based on hyperspectral technology [D].Xi'an:Northwest A & F University,2024:18-20.
[5] 杨涵,陈谦,王宝刚,等.利用高光谱技术预测采前猕猴桃干物质含量的可行性试验 [J].农业工程学报,2022,38(13):133-140.YANG H,CHEN Q,WANG B G,et al.Feasibility of estimating the dry matter content of kiwifruits before being harvested using hyperspectral technology [J].Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE ),2022,38(13):133-140.
[6] 张珮,王银红,江靖,等.便携式近红外光谱仪在果蔬品质定性和定量分析中的应用 [J].食品科技,2020,45(5):287-292.ZHANG P,WANG Y H,JIANG J,et al.Application of portable near-infrared spectrometer in qualitative and quantitative analysis of fruit and vegetable quality [J].Food Science and Technology,2020,45(5):287-292.
[7] 章恺,朱丽芳,李入林,等.基于改进 WOA-LSSVM 和高光谱的猕猴桃糖度无损检测 [J].食品与机械,2024,40(5):107-112,226.ZHANG K,ZHU L F,LI R L,et al.Non destructive detection of kiwifruit sugar content based on improved WOA-LSSVM and hyperspectral analysis [J].Food & Machinery,2024,40(5):107-112,226.
[8] 刘子涵,李明,赵峙尧,等.基于高光谱成像技术和机器学习的猕猴桃果实可溶性固形物含量预测 [J].果树学报,2024,41(12):2 606-2 620.LIU Z H,LI M,ZHAO S Y,et al.Prediction of soluble solids contents in kiwifruit based on both hyperspec-tral imaging technology and machine learning [J].Journal of Fruit Science,2024,41(12):2 606-2 620.
[9] 买买提·沙吾提,李荣鹏,蔡和兵,等.基于改进轻量级深度卷积神经网络的果树叶片分类及病害识别模型设计 [J].森林工程,2025,41(2):277-287.SAWUT M,LI R P,CAI H B,et al.Design of classification of fruit leaves and disease identification model based on improved MobileNet-V 2[J].Forest Engineering,2025,41(2):277-287.
[10] CHEN W H,CHEN H Z,FENG Q X,et al.A hybrid optimization method for sample partitioning in near-infrared analysis [J].Spectrochimica Acta Part A:Molecular and Biomolecular Spectroscopy,2021,248:119182.
[11] 刘畅,吴丹丹,王宁,等.基于二维相关红外光谱对 pH值影响大豆分离蛋白二级结构含量的快速分析 [J].食品科学,2024,45(17):26-34.LIU C,WU D D,WANG N,et al.Two-dimensional correlation infrared spectroscopy for rapid analysis of the effect of pH on the secondary structure content of soybean protein isolate [J].Food Science,2024,45(17):26-34.
[12] ZHENG X W,YANG C Y,ZHANG S K,et al.Ddpnas:efficient neural architecture search via dynamic distribution pruning [J].International Journal of Computer Vision,2023,131(5):1 234-1 249.
[13] RODRIGUEZ A,LAIO A.Clustering by fast search and find of density peaks [J].Science,2014,344(6 191):1 492-1 496.
[14] 梁子兆,李欣,刘朴,等.基于可见—近红外光谱的多品种猕猴桃贮藏品质的多指标综合预测模型研究 [J/OL ].食品工业科 技.(2024 -11-04) [2025 -02-16].https://doi.org/10.13386 /j.issn1002 -0306.2024070318.LIANG Z Z,LI X,LIU P,et al.Comprehensive multi-indicator prediction model for storage quality of multi-cultivar kiwifruit based on visible-near infrared spectroscopy [J/OL ].Science and Technology of Food Industry.(2024 -11-04) [2025 -02-16].https://doi.org/10.13386 /j.issn1002 -0306.2024070318.