ABSTRACT
Labour exploitation in the Taiwan Distant Water Fishing (DWF) industry has been a persistent issue for many years. Fishermen working on these vessels are often subjected to long working hours, low salaries, and poor living conditions. These conditions can lead to physical and mental health problems, exploitation, and abuse. To address this issue, a system has been developed with two modules. The first module collects data from three sources: CCTV footage from DWF vessels, Global fishing watch (GFW) open data, and Mobile Face Verification System (MFVS) interface for collecting data from fishermen and captains. The second module uses the collected data to identify and recognize instances of labour exploitation on DWF vessels. Our proposed system research shows that by combining different data sources, including MFVS, GFW, and Transfer-Learning, the You Only Look Once v7 (TL-YOLOv7) model can effectively identify and recognize labour exploitation. The proposed model aligns with Sustainable Development Goals by promoting decent work. It also improves working conditions to safeguard fishermen’s physical and mental health. The TL-YOLOv7 model achieves a higher mean average precision (mAP) value of 0.835 than the Pre-trained model of 0.691. This implies that the TL-YOLOv7 model exhibits higher accuracy in object detection. TL-YOLOv7 model achieves a lower RMSE of 0.44 compared to 5.24 for the GFW model, indicating a reduced overall deviation from the actual working hours. The system can help identify exploitation instances and promote better working conditions for fishermen in Taiwan’s DWF industry.
- Lauren Drakopulos, Jennifer J Silver, Eric Nost, Noella Gray, and Roberta Hawkins. 2022. Making global oceans governance in/visible with Smart Earth: The case of Global Fishing Watch. Environment and Planning E: Nature and Space (2022), 25148486221111786.Google Scholar
- Maria Gavouneli. 2019. Protecting Women Fishers: The Gender Parameters of Labour Rights at Sea. In Gender and the Law of the Sea. Brill Nijhoff, 165–179.Google Scholar
- Ross Girshick. 2015. Fast r-cnn. In Proceedings of the IEEE international conference on computer vision. 1440–1448.Google ScholarDigital Library
- Shih-Ming Kao. 2023. Analysis of Taiwan’s foreign fisheries policy in tuna RFMOs: Possible variables and determinants. Marine Policy 151 (2023), 105598.Google ScholarCross Ref
- P Karthikeyan, Chih Chun Chang, and Pao-Ann Hsiung. 2023. Labor exploitation investigation using statistical and multiple object tracking assessment methods. Multimedia Tools and Applications (2023), 1–24.Google Scholar
- P Karthikeyan and Pao-Ann Hsiung. 2022. Labour Exploitation Investigation using Satellite based Vessel Monitoring Systems. In 2022 3rd International Conference on Computing, Analytics and Networks (ICAN). IEEE, 1–5.Google ScholarCross Ref
- Christopher R Kerry, Owen M Exeter, and Matthew J Witt. 2022. Monitoring global fishing activity in proximity to seamounts using automatic identification systems. Fish and Fisheries 23, 3 (2022), 733–749.Google ScholarCross Ref
- GS Lee Son, S Romain, CS Rose, BJ Moore, KA Magrane, PS Packer, and FR Wallace. 2023. Development of electronic monitoring (EM) computer vision systems and machine learning algorithms for automated catch accounting in Alaska Fisheries. (2023).Google Scholar
- Chuyi Li, Lulu Li, Hongliang Jiang, Kaiheng Weng, Yifei Geng, Liang Li, Zaidan Ke, Qingyuan Li, Meng Cheng, Weiqiang Nie, 2022. YOLOv6: A single-stage object detection framework for industrial applications. arXiv preprint arXiv:2209.02976 (2022).Google Scholar
- Alejandro J Garcia Lozano, Jessica L Decker Sparks, Davina P Durgana, Courtney M Farthing, Juno Fitzpatrick, Birgitte Krough-Poulsen, Gavin McDonald, Sara McDonald, Yoshitaka Ota, Nicole Sarto, 2022. Decent work in fisheries: Current trends and key considerations for future research and policy. Marine Policy 136 (2022), 104922.Google ScholarCross Ref
- Gavin G McDonald, Christopher Costello, Jennifer Bone, Reniel B Cabral, Valerie Farabee, Timothy Hochberg, David Kroodsma, Tracey Mangin, Kyle C Meng, and Oliver Zahn. 2021. Satellites can reveal global extent of forced labor in the world’s fishing fleet. Proceedings of the National Academy of Sciences 118, 3 (2021), e2016238117.Google ScholarCross Ref
- Jaeyoon Park, Jennifer Van Osdel, Joanna Turner, Courtney M Farthing, Nathan A Miller, Hannah L Linder, Guillermo Ortuño Crespo, Gabrielle Carmine, and David A Kroodsma. 2023. Tracking elusive and shifting identities of the global fishing fleet. Science Advances 9, 3 (2023), eabp8200.Google Scholar
- Joseph Redmon, Santosh Divvala, Ross Girshick, and Ali Farhadi. 2016. You only look once: Unified, real-time object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition. 779–788.Google ScholarCross Ref
- Elizabeth R Selig, Shinnosuke Nakayama, Colette CC Wabnitz, Henrik Österblom, Jessica Spijkers, Nathan A Miller, Jan Bebbington, and Jessica L Decker Sparks. 2022. Revealing global risks of labor abuse and illegal, unreported, and unregulated fishing. Nature Communications 13, 1 (2022), 1612.Google ScholarCross Ref
- Katherine L Seto, Nathan A Miller, David Kroodsma, Quentin Hanich, Masanori Miyahara, Rui Saito, Kristina Boerder, Masaki Tsuda, Yoshioki Oozeki, and Osvaldo Urrutia S. 2023. Fishing through the cracks: The unregulated nature of global squid fisheries. Science advances 9, 10 (2023), eadd8125.Google Scholar
- Moritz Stäbler, Jonas Letschert, Marie Fujitani, and Stefan Partelow. 2022. Fish grabbing: Weak governance and productive waters are targets for distant water fishing. Plos one 17, 12 (2022), e0278481.Google ScholarCross Ref
- M Taconet, David Kroodsma, and JA Fernandes. 2019. Global atlas of AIS-based fishing activity—Challenges and opportunities. (2019).Google Scholar
- Hannah Thinyane and Michael Gallo. 2021. Negotiating trade-offs: Identifying labour exploitation in the fishing sector in Thailand. In ACM SIGCAS Conference on Computing and Sustainable Societies. 55–65.Google ScholarDigital Library
- Chien-Yao Wang, Alexey Bochkovskiy, and Hong-Yuan Mark Liao. 2022. YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors. arXiv preprint arXiv:2207.02696 (2022).Google Scholar
- Xingxing Xie, Gong Cheng, Jiabao Wang, Xiwen Yao, and Junwei Han. 2021. Oriented R-CNN for object detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision. 3520–3529.Google ScholarCross Ref
- Kuo-Wei Yen and Li-Chuan Liuhuang. 2021. A review of migrant labour rights protection in distant water fishing in Taiwan: From laissez-faire to regulation and challenges behind. Marine Policy 134 (2021), 104805.Google ScholarCross Ref
- Jinkai Yu and Qingchao Han. 2021. Exploring the management policy of distant water fisheries in China: Evolution, challenges and prospects. Fisheries Research 236 (2021), 105849.Google ScholarCross Ref
- Xingkui Zhu, Shuchang Lyu, Xu Wang, and Qi Zhao. 2021. TPH-YOLOv5: Improved YOLOv5 based on transformer prediction head for object detection on drone-captured scenarios. In Proceedings of the IEEE/CVF international conference on computer vision. 2778–2788.Google ScholarCross Ref
Index Terms
- Tackling Labour Exploitation in Taiwan Distant Water Fishing Industry through Automated Exploitation Detection System
Recommendations
Multimodal Labor Exploitation Detections for Taiwan Distant Water Fishing Industry
AbstractTaiwan plays a significant role in global seafood supply chains, accounting for approximately 10% of global tuna catches. The country is a substantial producer of tuna, shrimp, and squid, and its seafood industry is worth an estimated $1.3 billion ...
Labor exploitation investigation using statistical and multiple object tracking assessment methods
AbstractLabor exploitation in Taiwan’s Distant Water Fishing (DWF) industry has been a significant issue for many years. Migrant fishermen from Southeast Asian countries, including Indonesia and the Philippines, have reported being subjected to poor ...
Fishing Support System: Tying Fishing Line Automatically
ACIT '19: Proceedings of the 7th ACIS International Conference on Applied Computing and Information TechnologyThe current fishing population is getting small. As one of the causes, it is conceivable that young people in their 20s and 30s are getting away from fishing due to overuse smartphones. We thought that the population of fishing may increase by being ...
Comments