Abstract
[Objective] To enhance the accuracy of detecting fruit surface defects caused by minor mechanical friction during transportation. [Methods] A surface defect detection model for fused images of fruits based on improved ResNet 18 is proposed. To address the challenge of multi-scale feature extraction, the model incorporates a MELA module designed to capture image details more comprehensively. Additionally, an energy function-based regularization loss function is introduced to collaboratively optimize the training process alongside the cross-entropy loss function. Furthermore, a transfer learning strategy is employed to mitigate overfitting. [Results] Experimental results demonstrated that the improved model exhibits exceptional performance in fruit surface defect detection tasks, achieving the accuracy, precision, recall, and F1 score as high as 99.53%, 99.60%, 99.58%, and 99.58%, respectively. [Conclusion] The surface defect detection model for fused images of fruits based on improved ResNet 18 significantly enhances the detection accuracy and is suitable for surface defect detection in fused images of fruits.
Publication Date
7-25-2026
First Page
90
Last Page
97
DOI
10.13652/j.spjx.1003.5788.2025.80635
Recommended Citation
Haiping, Si; Wenrui, Zhao; Yuyang, Zhao; Tingting, Li; Bacao, Fernado; and Yanling, Li
(2026)
"A surface defect detection model for fused images of fruits based on improved ResNet 18,"
Food and Machinery: Vol. 42:
Iss.
7, Article 11.
DOI: 10.13652/j.spjx.1003.5788.2025.80635
Available at:
https://www.ifoodmm.cn/journal/vol42/iss7/11
References
[1] 吴中勇,李延荣,董中丹.我国水果市场发展现状及对策研究[J].中国果菜,2023,43(11):79-83,87.Wu Z Y,Li Y R,Dong Z D.Research on the current situation and countermeasures of fruit industry in China [J].China Fruit & Vegetable,2023,43(11):79-83,87.
[2] Meher B,Agrawal S,Panda R,et al.A survey on region based image fusion methods [J].Information Fusion,2019,48:119-132.
[3] Zhang X C,Demiris Y.Visible and infrared image fusion using deep learning [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2023,45(8):10 535-10 554.
[4] Li S T,Kang X D,Fang L Y,et al.Pixel-level image fusion:a survey of the state of the art [J].Information Fusion,2017,33:100-112.
[5] 褚璇,洪嘉隆,冯耿鑫,等.基于机器视觉的百香果品质多指标在线检测与分选 [J].食品与机械,2024,40(6):130-137,142.Chu X,Hong J L,Feng G X,et al.Online detection and sorting of passion fruit quality based on machine vision using multi-indicator [J].Food & Machinery,2024,40(6):130-137,142.
[6] Selvaraj A,Shebiah N,Nidhyanandhan S,et al.Fruit recognition using color and texture features [J].Journal of Emerging Trends in Computing and Information Sciences,2010,1(2):90-94.
[7] 王伟,杨世忠,宫钰程,等.EfficientNet V 2算法融合 GCN和CA-Transformer 的腐烂草莓分类方法 [J].食品与机械,2024,40(12):81-88.Wang W,Yang S Z,Gong Y C,et al.Rotten strawberry classification based on EfficientNet V 2 algorithm fused with GCN and CA-Transformer [J].Food & Machinery,2024,40(12):81-88.
[8] 王丹,刘芳菲,纪晓峰.基于双通道改进卷积神经网络的新鲜葡萄品质检测分析 [J].食品与机械,2024,40(12):89-94.Wang D,Liu F F,Ji X F.Fresh grape quality inspection based on dual-channel improved convolutional neural network [J].Food & Machinery,2024,40(12):89-94.
[9] Cai X J,Zhu Y,Liu S W,et al.FastSegFormer:a knowledge distillation-based method for real-time semantic segmentation of surface defects in navel oranges [J].Computers and Electronics in Agriculture,2024,217:108604.
[10] 侯文慧,郭丹丹,周传起,等.基于弱监督语义分割的砀山梨表面缺陷识别方法 [J].农业工程学报,2025,41(7):173-181.Hou W H,Guo D D,Zhou C X,et al.Recognizing the surface defects of Dangshan pears using weakly supervised semantic segmentation network [J].Transactions of the Chinese Society of Agricultural Engineering,2025,41(7):173-181.
[11] 王云鹏,司海平,宋佳珍,等.基于红外与可见光图像融合的苹果表面缺陷检测方法 [J].食品与机械,2021,37(12):127-131.Wang Y P,Si H P,Song J Z,et al.Apple surface defect detection method based on fusion of infrared and visible images [J].Food & Machinery,2021,37(12):127-131.
[12] Si H P,Wang Y P,Liu Q,et al.A registration algorithm for the infrared and visible images of apple based on active contour model [J].The Visual Computer,2024,40(4):2 833-2 855.
[13] 司海平,赵文汭,李婷婷,等.BDMFuse:基于红外与可见光图像基础特征和细节特征的多尺度融合 [J].红外与毫米波学报,2025,44(2):289-298.Si H P,Zhao W R,Li T T,et al.BDMFuse:multi-scale network fusion for infrared and visible images based on base and detail features [J].Journal of Infrared and Millimeter Waves,2025,44(2):289-298.[14] He K M,Zhang X Y,Ren S Q,et al.Deep residual learning for image recognition [C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR ).Las Vegas,NV:IEEE,2016:770-778.
[15] Xu W,Wan Y.ELA:efficient local attention for deep convolutional neural networks [J/OL ].Computer Science.(2024 -03-02) [2025 -07-11].https://arxiv.org/abs/ 2403.01123.
[16] Howard A G,Zhu M L,Chen B,et al.Mobilenets:efficient convolutional neural networks for mobile vision applications [J/OL].Computer Science.(2017 -04-17) [2025 -06-18].https://arxiv.org/abs/ 1704.04861.
[17] Yuan Y G,Xu B B,Hou L,et al.TEA:test-time energy adaptation [C]// 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR ).Seattle,WA:IEEE,2024:23 901-23 911.
[18] Krizhevsky A,Sutskever I,Hinton G E.ImageNet classification with deep convolutional neural networks [J].Communications of the ACM,2017,60(6):84-90.
[19] Simonyan K,Zisserman A.Very deep convolutional networks for large-scale image recognition [J/OL ].Computer Science.(2014 -11-04) [2025 -06-22].https://arxiv.org/abs/ 1409.1556.
[20] Tan M X,Le Q V.Efficientnetv 2:smaller models and faster training [J/OL ].Computer Science.(2021 -04-01) [2025 -07-08].https://arxiv.org/abs/ 2104.00298.
[21] Qin D F,Leichner C,Delakis M,et al.MobileNetV 4:universal models for the Mobile ecosystem [C]// Computer Vision-ECCV 2024.Cham:Springer Nature Switzerland,2025:78-96.
[22] Dosovitskiy A,Beyer L,Kolesnikov A,et al.An image is worth 16×16 words:transformers for image recognition at scale [J/OL ].Computer Science.(2020 -10-22) [2025 -07-02].https://arxiv.org/abs/ 2010.1129.
[23] Wang A,Chen H,Lin Z J,et al.Rep ViT:revisiting mobile CNN from ViT perspective [C]// 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR ).Seattle,WA:IEEE,2024:15 909-15 920.
[24] Zhu L H,Liao B C,Zhang Q,et al.Vision mamba:efficient visual representation learning with bidirectional state space model [J/OL ].Computer Science.(2024 -01-17) [2025 -06-29].https://arxiv.org/abs/ 2401.09417.
