文献类型：期刊文献

英文题名：Innovations in Seafood Freshness Quality: Non-Destructive Detection of Freshness in Litopenaeus Vannamei Using the Yolo-Shrimp Model

作者：Hou, Mingxin[1,2]; Zhong, Xiaowen[3]; Zheng, Ouyang[3]; Sun, Qinxiu[3]; Liu, Shucheng[3]; Liu, Mingxin[2,4]

机构：[1] School of Mechanical Engineering, Guangdong Ocean University, Guangdong Province, Zhanjiang, 524088, China; [2] Guangdong Provincial Key Laboratory of Intelligent Equipment for South China Sea Marine Ranching, Guangdong Ocean University, Zhanjiang, 524088, China; [3] College of Food Science and Technology, Guangdong Ocean University, Guangdong Province, Zhanjiang, 524088, China; [4] School of Electronics and Information Engineering, Guangdong Ocean University, Guangdong Province, Zhanjiang, 524088, China

年份：2024

外文期刊名：SSRN

收录：EI(收录号：20240221916)

语种：英文

外文关键词：Biochemistry - Safety engineering - Textures

外文摘要：In this work, changes in the quality of Litopenaeus vannamei stored at 4 °C for one week are investigated, focusing on biochemical markers such as total volatile basic nitrogen (TVB-N), total viable count (TVC), and the degree of melanosis, along with their correlations. Additionally, the YOLO-Shrimp model, an advanced version of the YOLOv8 architecture incorporating the focal EIOU loss function and the C3X computation module, is introduced. These enhancements significantly improve the precision and adaptability of the model for assessing shrimp freshness. The YOLO-Shrimp model employs a non-destructive, rapid assessment method by analyzing shrimp texture, color and morphological attributes. Compared to YOLOv8, performance improvements were observed in precision (5.07%), recall (1.58%), F1 score (3.25%), and mAP50 (2.84%). Empirical validations confirmed that the model's assessments align with the biochemical, microbiological, and physical indicators, highlighting its effectiveness in detecting shrimp freshness and its potential to enhance food safety and quality control standards. ? 2024, The Authors. All rights reserved.