{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2024,7,30]],"date-time":"2024-07-30T08:40:28Z","timestamp":1722328828586},"reference-count":18,"publisher":"National Library of Serbia","issue":"2","license":[{"start":{"date-parts":[[2024,1,1]],"date-time":"2024-01-01T00:00:00Z","timestamp":1704067200000},"content-version":"unspecified","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["ComSIS","COMPUT SCI INF SYST","COMPUT SCI INFORM SY","COMPUTER SCI INFORM","COMSIS J"],"published-print":{"date-parts":[[2024]]},"abstract":"<jats:p>When data collection is limited, such as in the case of fire detection, improving the detection rate with only number of small labeled data is difficult. Therefore, researchers have conducted many related studies, among which semisupervised learning methods have achieved good results in improving detection rates. Most recent semi-supervised learning models use the pseudo-label method. But there is a problem, which is that it is difficult to label accurately in samples that deviate from the true label distribution due to false labels. In other words, due to the pseudo-label used for data augmentation, erroneous biases can be accumulated and adversely affect the final weights. To improve this, we proposed a method of generating Similar-labeled data (prediction result labeling value and correct answer value are similar), which was used through the F-guessed method and the Region of Interest (ROI) expression method in the video during initial learning. This has the effect of preventing the bias from being distorted in the initial stages. As a result, data generation increased by about 6.5 times, from 5,565 to 41,712, mAP@0.5 increased by about 26.1%, from 65.9% to 92.0%, and loss improved from 3.347 to 1.69, compared to the initial labeled data.<\/jats:p>","DOI":"10.2298\/csis230820011k","type":"journal-article","created":{"date-parts":[[2024,4,30]],"date-time":"2024-04-30T11:12:28Z","timestamp":1714475548000},"page":"645-661","source":"Crossref","is-referenced-by-count":0,"title":["A study on fire data augmentation from video\/image using the similar-label and F-guessed method"],"prefix":"10.2298","volume":"21","author":[{"given":"Jong-Sik","family":"Kim","sequence":"first","affiliation":[{"name":"Dept. of Electronics Engineering, Dong-A University, Nakdong-daero beon-gil Saha-gu, Busan, Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dae-Seong","family":"Kang","sequence":"additional","affiliation":[{"name":"Dept. of Electronics Engineering, Dong-A University, Nakdong-daero beon-gil Saha-gu, Busan, Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1078","reference":[{"key":"ref1","unstructured":"Amit Chaudhary.: Semi-Supervised Learning in Computer Vision\u201d, https:\/\/amitness.com\/2020\/07\/semi-supervised-learning [accessed: Sep. 10, 2022]"},{"key":"ref2","unstructured":"Yassine Ouali, C\u00b4eline Hudelot, and Myriam Tami.: An Overview of Deep Semi-Supervised Learning, Machine Learning (cs.LG), arXiv:2006.05278, Jul. 2020."},{"key":"ref3","unstructured":"Dong-Hyun Lee.: Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks\u201d, In ICMLW, 2013."},{"key":"ref4","unstructured":"Vinko Kod\u02c7zoman: Pseudo-labeling a simple semi-supervised learning method, https:\/\/datawhatnow.com\/pseudo-labeling-semi-supervised-learning [accessed: Apr. 10, 2023]"},{"key":"ref5","doi-asserted-by":"crossref","unstructured":"Hieu Pham, Zihang Dai, Qizhe Xie and Quoc V. Le.: Meta Pseudo Labels, Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 11557- 11568, 2021.","DOI":"10.1109\/CVPR46437.2021.01139"},{"key":"ref6","unstructured":"Baixu Chen, Junguang Jiang, XimeiWang, PengfeiWan, JianminWang, and Mingsheng Long: Debiased Self-Training for Semi-Supervised Learning, Advances in Neural Information Processing Systems 35 (NeurIPS 2022), arXiv:2202.07136v5, Nov 2022."},{"key":"ref7","unstructured":"Barret Zoph, Golnaz Ghiasi, Tsung-Yi Lin, Yin Cui, Hanxiao Liu, Ekin Dogus Cubuk, and Quoc Le.: Rethinking Pre-training and Self-training, Neural Information Processing Systems 33, 2020."},{"key":"ref8","unstructured":"Mengde Xu, Zheng Zhang, Han Hu, Jianfeng Wang, Lijuan Wang, Fangyun Wei, Xiang Bai, and Zicheng Liu.: End-to-End Semi-Supervised Object Detection with Soft Teache, IEEE\/CVF International Conference on Computer Vision (ICCV), pp. 3060-3069, 2021."},{"key":"ref9","doi-asserted-by":"crossref","unstructured":"Xiaokang Chen, Yuhui Yuan, Gang Zeng, and Jingdong Wang.: Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision, Computer Vision and Pattern Recognition (CVPR), pp. 2613-2622, Jun. 2021.","DOI":"10.1109\/CVPR46437.2021.00264"},{"key":"ref10","unstructured":"David Berthelot, Nicholas Carlini, Ian Goodfellow, Nicolas Papernot, Avital Oliver, and Colin A Raffel.: MixMatch: A Holistic Approach to Semi-Supervised Learning, Neural Information Processing Systems 32, 2019."},{"key":"ref11","unstructured":"David Berthelot, Nicholas Carlini, Ekin D. Cubuk, Alex Kurakin, Kihyuk Sohn, Han Zhang and Colin Raffel.: ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring, Machine Learning (stat.ML), arXiv:1911.09785, Feb. 2020."},{"key":"ref12","unstructured":"Kihyuk Sohn, David Berthelot, Nicholas Carlini, Zizhao Zhang, Han Zhang, Colin A. Raffel, Ekin Dogus Cubuk, Alexey Kurakin, and Chun-Liang Li.: FixMatch: Simplifying Semi- Supervised Learning with Consistency and Confidence, Neural Information Processing Systems 33, 2020."},{"key":"ref13","unstructured":"Ananya Kumar, Aditi Raghunathan, Robbie Jones, Tengyu Ma, and Percy Liang.: Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distribution, Computer Vision and Pattern Recognition(cs.CV), arXiv:2202.10054, Feb. 2022."},{"key":"ref14","unstructured":"Marius Mosbach, Maksym Andriushchenko, and Dietrich Klakow.: On the Stability of Fine-tuning BERT: Misconceptions, Explanations, and Strong Baselines, Machine Learning (stat.ML), arXiv:2006. 04884, Mar. 2021."},{"key":"ref15","doi-asserted-by":"crossref","unstructured":"Jong-Sik Kim and Dae-Seong Kang.: A Study on Fire Data Generation and Recognition Rate Improvement using F-guessed and Semi-supervised Learning, The Journal of Korean Institute of Information Technology, Vol. 20, pp. 123-134, Dec 2022.","DOI":"10.14801\/jkiit.2022.20.12.123"},{"key":"ref16","unstructured":"Alexey Bochkovskiy, Chien-Yao Wang, Hong- Yuan, and Mark Liao.: YOLOv4: Optimal Speed and Accuracy of Object Detection, Computer Vision and Pattern Recognition (cs.CV), arXiv:2004. 10934, Apr. 2020."},{"key":"ref17","unstructured":"AI Hub data, https:\/\/aihub.or.kr [accessed: Apr. 10, 2023]"},{"key":"ref18","doi-asserted-by":"crossref","unstructured":"Hye-Youn Lim, Jun-Mock Lee and Dae-Seong Kang.: A Method for Improving Learning Convergence Curve and Learning Time of DA-FSL Model using Knowledge Distillation, The Journal of Korean Institute of Information Technology, Vol. 18, pp. 25-32, Oct. 2020.","DOI":"10.14801\/jkiit.2020.18.10.25"}],"container-title":["Computer Science and Information Systems"],"original-title":[],"language":"en","deposited":{"date-parts":[[2024,7,30]],"date-time":"2024-07-30T08:22:53Z","timestamp":1722327773000},"score":1,"resource":{"primary":{"URL":"https:\/\/doiserbia.nb.rs\/Article.aspx?ID=1820-02142400011K"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024]]},"references-count":18,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2024]]}},"URL":"https:\/\/doi.org\/10.2298\/csis230820011k","relation":{},"ISSN":["1820-0214","2406-1018"],"issn-type":[{"type":"print","value":"1820-0214"},{"type":"electronic","value":"2406-1018"}],"subject":[],"published":{"date-parts":[[2024]]}}}