{"title":"MFSnet:用于密集环境下鸡计数的多尺度特征筛选网络。","authors":"G Ma, Z Xiao, F Yuan, E Sun, S Chen, J Liu, B He","doi":"10.1080/00071668.2025.2500346","DOIUrl":null,"url":null,"abstract":"<p><p>1. Machine-vision-based chicken counting is a highly efficient approach. Nonetheless, in scenarios with high breeding densities, chickens in the captured images frequently overlap with one another. This research addressed the challenge of accurately counting chickens within a free-range chicken coop in densely environments. It proposes a chicken-counting network specifically designed for dense scenarios, namely MFSnet.2. The study extracted multi-scale feature maps and subjected them to processing during the fusion stage via a Feature Screening Module (FSM). This module generated feature maps that were richly endowed with features from diverse scales to enhance information, thereby augmenting the network's capacity to accurately identify chickens.3. The dataset was collected and labelled and denominated as Chicken2023. It consisted of 550 images, which, in aggregate, encompassed a total of 49 747 chickens. To validate its efficacy, it was compared with extant counting algorithms. The experimental findings derived from the Chicken2023 dataset illustrated that this method attained a better counting performance level. It achieved a mean absolute error (MAE) of 2.7 and a root mean square error (RMSE) of 3.6. When juxtaposed with the top-performing network, it showed a notable improvement, with a 6.25% reduction in MAE and a 6.26% reduction in RMSE.4. The network model proposed in this study accurately recognised the number of chickens in dense environments and improved the efficiency of poultry farming.</p>","PeriodicalId":9322,"journal":{"name":"British Poultry Science","volume":" ","pages":"1-11"},"PeriodicalIF":1.7000,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"MFSnet: a multi-scale feature screening network for chicken counting in dense environments.\",\"authors\":\"G Ma, Z Xiao, F Yuan, E Sun, S Chen, J Liu, B He\",\"doi\":\"10.1080/00071668.2025.2500346\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>1. Machine-vision-based chicken counting is a highly efficient approach. Nonetheless, in scenarios with high breeding densities, chickens in the captured images frequently overlap with one another. This research addressed the challenge of accurately counting chickens within a free-range chicken coop in densely environments. It proposes a chicken-counting network specifically designed for dense scenarios, namely MFSnet.2. The study extracted multi-scale feature maps and subjected them to processing during the fusion stage via a Feature Screening Module (FSM). This module generated feature maps that were richly endowed with features from diverse scales to enhance information, thereby augmenting the network's capacity to accurately identify chickens.3. The dataset was collected and labelled and denominated as Chicken2023. It consisted of 550 images, which, in aggregate, encompassed a total of 49 747 chickens. To validate its efficacy, it was compared with extant counting algorithms. The experimental findings derived from the Chicken2023 dataset illustrated that this method attained a better counting performance level. It achieved a mean absolute error (MAE) of 2.7 and a root mean square error (RMSE) of 3.6. When juxtaposed with the top-performing network, it showed a notable improvement, with a 6.25% reduction in MAE and a 6.26% reduction in RMSE.4. The network model proposed in this study accurately recognised the number of chickens in dense environments and improved the efficiency of poultry farming.</p>\",\"PeriodicalId\":9322,\"journal\":{\"name\":\"British Poultry Science\",\"volume\":\" \",\"pages\":\"1-11\"},\"PeriodicalIF\":1.7000,\"publicationDate\":\"2025-06-02\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"British Poultry Science\",\"FirstCategoryId\":\"97\",\"ListUrlMain\":\"https://doi.org/10.1080/00071668.2025.2500346\",\"RegionNum\":3,\"RegionCategory\":\"农林科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"AGRICULTURE, DAIRY & ANIMAL SCIENCE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"British Poultry Science","FirstCategoryId":"97","ListUrlMain":"https://doi.org/10.1080/00071668.2025.2500346","RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"AGRICULTURE, DAIRY & ANIMAL SCIENCE","Score":null,"Total":0}
MFSnet: a multi-scale feature screening network for chicken counting in dense environments.
1. Machine-vision-based chicken counting is a highly efficient approach. Nonetheless, in scenarios with high breeding densities, chickens in the captured images frequently overlap with one another. This research addressed the challenge of accurately counting chickens within a free-range chicken coop in densely environments. It proposes a chicken-counting network specifically designed for dense scenarios, namely MFSnet.2. The study extracted multi-scale feature maps and subjected them to processing during the fusion stage via a Feature Screening Module (FSM). This module generated feature maps that were richly endowed with features from diverse scales to enhance information, thereby augmenting the network's capacity to accurately identify chickens.3. The dataset was collected and labelled and denominated as Chicken2023. It consisted of 550 images, which, in aggregate, encompassed a total of 49 747 chickens. To validate its efficacy, it was compared with extant counting algorithms. The experimental findings derived from the Chicken2023 dataset illustrated that this method attained a better counting performance level. It achieved a mean absolute error (MAE) of 2.7 and a root mean square error (RMSE) of 3.6. When juxtaposed with the top-performing network, it showed a notable improvement, with a 6.25% reduction in MAE and a 6.26% reduction in RMSE.4. The network model proposed in this study accurately recognised the number of chickens in dense environments and improved the efficiency of poultry farming.
期刊介绍:
From its first volume in 1960, British Poultry Science has been a leading international journal for poultry scientists and advisers to the poultry industry throughout the world. Over 60% of the independently refereed papers published originate outside the UK. Most typically they report the results of biological studies with an experimental approach which either make an original contribution to fundamental science or are of obvious application to the industry. Subjects which are covered include: anatomy, embryology, biochemistry, biophysics, physiology, reproduction and genetics, behaviour, microbiology, endocrinology, nutrition, environmental science, food science, feeding stuffs and feeding, management and housing welfare, breeding, hatching, poultry meat and egg yields and quality.Papers that adopt a modelling approach or describe the scientific background to new equipment or apparatus directly relevant to the industry are also published. The journal also features rapid publication of Short Communications. Summaries of papers presented at the Spring Meeting of the UK Branch of the WPSA are published in British Poultry Abstracts .