Line-labelling enhanced CNNs for transparent juvenile fish crowd counting
Counting juvenile fish in aquaculture is challenging due to their small, fragile, and often transparent bodies, especially under high-density conditions. To address this, we propose a novel line-labeling annotation method specifically designed for transparent juvenile fish counting, which enhances s...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-08-01
|
| Series: | Smart Agricultural Technology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2772375525001960 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850259016456863744 |
|---|---|
| author | Dianzhuo Zhou Hequn Tan Yuxiang Li Yuxuan Deng Ming Zhu |
| author_facet | Dianzhuo Zhou Hequn Tan Yuxiang Li Yuxuan Deng Ming Zhu |
| author_sort | Dianzhuo Zhou |
| collection | DOAJ |
| description | Counting juvenile fish in aquaculture is challenging due to their small, fragile, and often transparent bodies, especially under high-density conditions. To address this, we propose a novel line-labeling annotation method specifically designed for transparent juvenile fish counting, which enhances supervision quality and provides both positional and morphological cues. We also introduce an improved CSRNet-based convolutional neural network, optimized for high-density fish scenarios. A dataset of 9000 annotated images of Silver Carp and Tilapia, categorized into four density ranges (0–10, 10–20, 20–30 and 30–40 fish/cm²), was used to train and evaluate our method. To determine the optimal approach, four combinations of labeling and image enhancement methods were tested: Point Labeling + Original Image (P + O), Line Labeling + Original Image (L + O), Point Labeling + Image Enhancement (P + I) and Line Labeling + Image Enhancement (L + I). Counting accuracy was assessed using heatmap-based visualizations. Experimental results demonstrate that the line-labeling method significantly improves counting accuracy, achieving 97.73 % for Silver Carp and 98.04 % for Tilapia, outperforming conventional point-based annotations in high-density contexts. This study highlights the potential of structured annotations and tailored network designs for advancing precision in fish counting tasks. |
| format | Article |
| id | doaj-art-b0d8a674582c4d3f998fa4a6ac0460b5 |
| institution | OA Journals |
| issn | 2772-3755 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Smart Agricultural Technology |
| spelling | doaj-art-b0d8a674582c4d3f998fa4a6ac0460b52025-08-20T01:55:58ZengElsevierSmart Agricultural Technology2772-37552025-08-011110096310.1016/j.atech.2025.100963Line-labelling enhanced CNNs for transparent juvenile fish crowd countingDianzhuo Zhou0Hequn Tan1Yuxiang Li2Yuxuan Deng3Ming Zhu4Key Laboratory of Aquaculture Facilities Engineering, Ministry of Agriculture and Rural Affairs, Wuhan 430070, China; College of Engineering, Huazhong Agricultural University, Wuhan 430070, ChinaKey Laboratory of Aquaculture Facilities Engineering, Ministry of Agriculture and Rural Affairs, Wuhan 430070, China; College of Engineering, Huazhong Agricultural University, Wuhan 430070, China; Corresponding author at: No.1 Shizishan Street Hongshan District, Wuhan, 430070, China.Key Laboratory of Aquaculture Facilities Engineering, Ministry of Agriculture and Rural Affairs, Wuhan 430070, China; College of Engineering, Huazhong Agricultural University, Wuhan 430070, ChinaKey Laboratory of Aquaculture Facilities Engineering, Ministry of Agriculture and Rural Affairs, Wuhan 430070, China; College of Engineering, Huazhong Agricultural University, Wuhan 430070, ChinaKey Laboratory of Aquaculture Facilities Engineering, Ministry of Agriculture and Rural Affairs, Wuhan 430070, China; College of Engineering, Huazhong Agricultural University, Wuhan 430070, ChinaCounting juvenile fish in aquaculture is challenging due to their small, fragile, and often transparent bodies, especially under high-density conditions. To address this, we propose a novel line-labeling annotation method specifically designed for transparent juvenile fish counting, which enhances supervision quality and provides both positional and morphological cues. We also introduce an improved CSRNet-based convolutional neural network, optimized for high-density fish scenarios. A dataset of 9000 annotated images of Silver Carp and Tilapia, categorized into four density ranges (0–10, 10–20, 20–30 and 30–40 fish/cm²), was used to train and evaluate our method. To determine the optimal approach, four combinations of labeling and image enhancement methods were tested: Point Labeling + Original Image (P + O), Line Labeling + Original Image (L + O), Point Labeling + Image Enhancement (P + I) and Line Labeling + Image Enhancement (L + I). Counting accuracy was assessed using heatmap-based visualizations. Experimental results demonstrate that the line-labeling method significantly improves counting accuracy, achieving 97.73 % for Silver Carp and 98.04 % for Tilapia, outperforming conventional point-based annotations in high-density contexts. This study highlights the potential of structured annotations and tailored network designs for advancing precision in fish counting tasks.http://www.sciencedirect.com/science/article/pii/S2772375525001960Juvenile fish countingPoint labelingLine labelingImage enhancementHeatmapFish density |
| spellingShingle | Dianzhuo Zhou Hequn Tan Yuxiang Li Yuxuan Deng Ming Zhu Line-labelling enhanced CNNs for transparent juvenile fish crowd counting Smart Agricultural Technology Juvenile fish counting Point labeling Line labeling Image enhancement Heatmap Fish density |
| title | Line-labelling enhanced CNNs for transparent juvenile fish crowd counting |
| title_full | Line-labelling enhanced CNNs for transparent juvenile fish crowd counting |
| title_fullStr | Line-labelling enhanced CNNs for transparent juvenile fish crowd counting |
| title_full_unstemmed | Line-labelling enhanced CNNs for transparent juvenile fish crowd counting |
| title_short | Line-labelling enhanced CNNs for transparent juvenile fish crowd counting |
| title_sort | line labelling enhanced cnns for transparent juvenile fish crowd counting |
| topic | Juvenile fish counting Point labeling Line labeling Image enhancement Heatmap Fish density |
| url | http://www.sciencedirect.com/science/article/pii/S2772375525001960 |
| work_keys_str_mv | AT dianzhuozhou linelabellingenhancedcnnsfortransparentjuvenilefishcrowdcounting AT hequntan linelabellingenhancedcnnsfortransparentjuvenilefishcrowdcounting AT yuxiangli linelabellingenhancedcnnsfortransparentjuvenilefishcrowdcounting AT yuxuandeng linelabellingenhancedcnnsfortransparentjuvenilefishcrowdcounting AT mingzhu linelabellingenhancedcnnsfortransparentjuvenilefishcrowdcounting |