Academic Journal
YOLO-MCD: An Efficient Detection Method for Multi-Dish Recognition.
| Τίτλος: | YOLO-MCD: An Efficient Detection Method for Multi-Dish Recognition. |
|---|---|
| Συγγραφείς: | Yu S; College of Mathematics and Computer Science, Zhejiang A&F University, Lin'an, Hangzhou, Zhejiang, China., Yang Y; College of Mathematics and Computer Science, Zhejiang A&F University, Lin'an, Hangzhou, Zhejiang, China., Gu X; Tianmu Mountain Forest Farm, Lin'an, Hangzhou, Zhejiang, China., Zhu T; College of Mathematics and Computer Science, Zhejiang A&F University, Lin'an, Hangzhou, Zhejiang, China., Xu T; Anxiao Technology (Hangzhou) Co. Ltd, Hangzhou, Zhejiang, China. |
| Πηγή: | Journal of food science [J Food Sci] 2026 Sep; Vol. 91 (9), pp. e71484. |
| Τύπος έκδοσης: | Journal Article |
| Γλώσσα: | English |
| Στοιχεία περιοδικού: | Publisher: Wiley on behalf of the Institute of Food Technologists Country of Publication: United States NLM ID: 0014052 Publication Model: Print Cited Medium: Internet ISSN: 1750-3841 (Electronic) Linking ISSN: 00221147 NLM ISO Abbreviation: J Food Sci Subsets: MEDLINE |
| Imprint Name(s): | Publication: Malden, Mass. : Wiley on behalf of the Institute of Food Technologists Original Publication: Champaign, Ill. Institute of Food Technologists |
| Ιατρικοί όροι (MeSH): | Food Analysis*/methods , Detection Algorithms* |
| Περίληψη: | In response to the challenges of insufficient detection accuracy, limited processing speed, and poor recognition of long-tail categories in intelligent catering scenarios, this paper proposes YOLO-MCD (YOLO-multi-Chinese-dish), a fast and accurate multi-dish detection method. Based on YOLO11, YOLO-MCD introduces three key improvements: the design of the C3K2_RFCBAMConv module to enhance feature extraction capability; the incorporation of an EUCB upsampling structure to strengthen multi-scale feature fusion; and the construction of a dynamic loss function EMASlideLoss, which integrates the exponential moving average with an IoU-based adaptive weighting mechanism to significantly improve the detection performance of long-tail categories. Experiments on a self-built multi-dish dataset demonstrate that YOLO-MCD achieves 91.2% mAP@50 and 86.4% mAP@50-95, with an inference speed of 278 FPS, outperforming YOLO11n by 2.2% in mAP@50 while increasing the parameter count by only 7%. Compared with SSD, Faster R-CNN, and RT-DETR-R18, YOLO-MCD exhibits superior performance in both accuracy and speed. Furthermore, ablation studies verify the independent effectiveness and synergistic gains of each module. YOLO-MCD maintains real-time inference capability while achieving high-precision detection for long-tail distributions, multi-object scenes, and complex backgrounds, demonstrating great potential for applications in intelligent catering and food safety. (© 2026 Institute of Food Technologists.) |
| References: | Comput Intell Neurosci. 2021 Dec 16;2021:1268453. (PMID: 34956342) IEEE Trans Image Process. 2021;30:1514-1526. (PMID: 33360994) Food Chem. 2025 Feb 1;464(Pt 2):141739. (PMID: 39461309) IEEE J Biomed Health Inform. 2017 May;21(3):588-598. (PMID: 28114043) Food Chem. 2024 Feb 15;434:137525. (PMID: 37742550) IEEE J Biomed Health Inform. 2016 May;20(3):848-855. (PMID: 25850095) Foods. 2025 Jan 31;14(3):. (PMID: 39942054) Food Res Int. 2025 Feb;201:115675. (PMID: 39849794) IEEE Trans Pattern Anal Mach Intell. 2023 Aug;45(8):9932-9949. (PMID: 37021867) Front Nutr. 2022 Nov 16;9:965801. (PMID: 36466396) Food Res Int. 2021 Sep;147:110437. (PMID: 34399450) IEEE J Biomed Health Inform. 2014 Jul;18(4):1261-71. (PMID: 25014934) |
| Grant Information: | LQ20F020005 Zhejiang Provincial Natural Science Foundation of China |
| Contributed Indexing: | Keywords: feature enhancement; intelligent catering; long‐tail distribution; multi‐dish detection; real‐time detection; yolo11 |
| Entry Date(s): | Date Created: 20260914 Date Completed: 20260914 Latest Revision: 20260916 |
| Update Code: | 20260916 |
| PubMed Central ID: | PMC13572774 |
| DOI: | 10.1111/1750-3841.71484 |
| PMID: | 42733153 |
| Βάση Δεδομένων: | MEDLINE |
| ISSN: | 1750-3841 |
|---|---|
| DOI: | 10.1111/1750-3841.71484 |