관개용수로 CCTV 이미지를 이용한 CNN 딥러닝 이미지 모델 적용
Application of CCTV Image and Semantic Segmentation Model for Water Level Estimation of Irrigation Channel
0
조회
0
다운로드
4
인용
0
참고문헌
키워드
초록
A more accurate understanding of the irrigation water supply is necessary for efficient agricultural water management. Although we measure water levelsin an irrigation canal using ultrasonic water level gauges, some errors occur due to malfunctions or the surrounding environment. This study aims toapply CNN (Convolutional Neural Network) Deep-learning-based image classification and segmentation models to the irrigation canal’s CCTV(Closed-Circuit Television) images. The CCTV images were acquired from the irrigation canal of the agricultural reservoir in Cheorwon-gun,Gangwon-do. We used the ResNet-50 model for the image classification model and the U-Net model for the image segmentation model. Using theNatural Breaks algorithm, we divided water level data into 2, 4, and 8 groups for image classification models. The classification models of 2, 4, and8 groups showed the accuracy of 1.000, 0.987, and 0.634, respectively. The image segmentation model showed a Dice score of 0.998 and predictedwater levels showed R2of 0.97 and MAE (Mean Absolute Error) of 0.02 m. The image classification models can be applied to the automaticgate-controller at four divisions of water levels. Also, the image segmentation model results can be applied to the alternative measurement for ultrasonicwater gauges. We expect that the results of this study can provide a more scientific and efficient approach for agricultural water management.
Abstract
A more accurate understanding of the irrigation water supply is necessary for efficient agricultural water management. Although we measure water levelsin an irrigation canal using ultrasonic water level gauges, some errors occur due to malfunctions or the surrounding environment. This study aims toapply CNN (Convolutional Neural Network) Deep-learning-based image classification and segmentation models to the irrigation canal’s CCTV(Closed-Circuit Television) images. The CCTV images were acquired from the irrigation canal of the agricultural reservoir in Cheorwon-gun,Gangwon-do. We used the ResNet-50 model for the image classification model and the U-Net model for the image segmentation model. Using theNatural Breaks algorithm, we divided water level data into 2, 4, and 8 groups for image classification models. The classification models of 2, 4, and8 groups showed the accuracy of 1.000, 0.987, and 0.634, respectively. The image segmentation model showed a Dice score of 0.998 and predictedwater levels showed R2of 0.97 and MAE (Mean Absolute Error) of 0.02 m. The image classification models can be applied to the automaticgate-controller at four divisions of water levels. Also, the image segmentation model results can be applied to the alternative measurement for ultrasonicwater gauges. We expect that the results of this study can provide a more scientific and efficient approach for agricultural water management.
참고문헌 (21)
- 방재홍, "관개용수로의 자동수위측정 자료를 활용한 농업용 저수지 공급량 산정 및 분석", 한국농공학회논문집, 63(1), pp.27~35 (2021), 한국농공학회[DOI]
- Cao, Y., "HENet : head-level ensemble network for very high resolution remote sensing images semantic segmentation", IEEE Geoscience and Remote Sensing Letters, 19, pp.1~5 (2022)[DOI]
- Flood-water level estimation from social media images, "Chau dhary, P.", The International Archives of the Photogrammetry, 5194~, pp.Remote Sensing and Spatial Information Sciences Ⅳ-2 (2019), W5
- Development of automatic flow measurement technology to enhance disaster-safety codes for small stream, "Cheong, T. S.", National Disaster Management Research Institute, 98~101 (2018)
- Binge Cui, "Semantic Segmentation of Remote Sensing Images Using Transfer Learning and Deep Convolutional Neural Network With Dense Connection", IEEE Access, 8, pp.116744~116755 (2020), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- He, K., "Deep residual learning for image recognition", arXiv: 1512.03385v1, ~ (2015)
- 홍은미, "실시간 관개수로 수위 모니터링을 활용한 논 관개용수 공급적정성 평가", 한국농공학회논문집, 56(4), pp.1~8 (2014), 한국농공학회
- "2021", ~
- 김규전, "스테레오 CCTV 시스템을 이용한 수위측정 정확도 향상", Crisisonomy, 15(4), pp.83~93 (2019), (사)위기관리이론과실천[DOI]
- 김선주, "농업용수 정량화를 위한 경제적 수위계측망 설계", 한국농공학회논문집, 58(5), pp.19~28 (2016), 한국농공학회
- Kingma, D. P., "Adam : a method for stochastic optimization", arXiv:1412.6980v9, ~ (2014)
- E Mique, "Deep Residual U-Net Based Lung Image Segmentation for Lung Disease Detection", IOP Conference Series: Materials Science and Engineering, 803(1), pp.012004~ (2020), IOP Publishing[DOI]
- Maehara, H., "Water level measurement from CCTV camera images using water gauge images taken at the time of low water level", Journal of the Japan Society of Photogrammetry and Remote Sensing, 55(1), pp.66~68 (2016)[DOI]
- Hideaki MAEHARA, "A deep-learning based water-level measurement method from CCTV camera images", Journal of the Japan society of photogrammetry and remote sensing, 58(1), pp.28~33 (2019), Japan Society of Photogrammetry and Remote Sensing[DOI]
- 이재남, "수로부 계측정보 기반 농업용 저수지의 관개용수 공급량 평가", 한국농공학회논문집, 62(6), pp.63~72 (2020), 한국농공학회[DOI]
- Jingyu Li, "Lane-DeepLab: Lane semantic segmentation in automatic driving scenarios for high-definition maps", Neurocomputing, 465, pp.15~25 (2021), Elsevier BV[DOI]
- Yan-Ting Lin, "Automatic water-level detection using single-camera images with varied poses", Measurement, 127, pp.167~174 (2018), Elsevier BV[DOI]
- Laura Lopez-Fuentes, "River segmentation for flood monitoring", 2017 IEEE International Conference on Big Data (Big Data), pp.3760~3763 (2017), IEEE
- "2019", ~
- Pedro Vianna, "U-Net and SegNet performances on lesion segmentation of breast ultrasonography images", Research on Biomedical Engineering, 37(2), pp.171~179 (2021), Springer Science and Business Media LLC[DOI]
- Yuan X., "A review of deep learning methods for semantic segmentation of remote sensing imagery", Expert Systems with Applications, 169, pp.114417~ (2021)[DOI]