수수 재배지 분할을 위한 무인기 취득 RGB 영상 분석
Analysis of RGB Imagery Acquired by UAVs for Sorghum Field Segmentation
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
본 연구는 수수를 대상으로 수수 주산지에서 취득한 무인기 기반 RGB영상에 Seg-Net과 U-Net모델을 작성 후, 일반화 가능성을 검토하여 실시간 재배지 파악에 더 효율적인 기법을 제안하기 위해 수행되었다. 경상북도 안동시의 수수 재배지 영상 264장을 이용해 모델학습을 진행하고, 충청북도 제천시 수수 재배지 영상 14장을 이용하여 테스트를 진행하였다. U-Net의 학습은 14epoch, AccuracyC = 0.9426, LossC = 0.1593, Dice_coefC = 0.9223, AccuracyV = 0.6403, LossV = 1.9624, Dice_coefV = 0.6402에 4시간 37분이 소요되었으며, Seg-Net의 학습은 101epoch, AccuracyC = 0.6363, LossC = 0.6573, Dice_coefC = 0.5586, AccuracyV = 0.5711, LossV = 0.6785, Dice_coefV = 0.5586이며 1시간 55분이 소요되었다. Test결과 U-Net은 AccuracyT = 0.6806, LossT = 0.7180, Dice_coefT = 0.5558, Seg-Net은 AccuracyT = 0.7472, LossT = 0.5225, Dice_coefC = 0.6159로 나타났다. Seg-Net의 Calibration성능은 낮지만 일반화 성능이 뛰어나며, 모델의 빠른 학습 시간, 더 낮은 메모리 요구량으로 수수 재배지 분할에서 U-Net보다 효율적인 모델이라고 사료된다.
Abstract
This study was conducted to propose a more efficient method for real-time cultivation area detection by examining the generalization capability of Seg-Net and U-Net models using UAV-based RGB imagery obtained from major sorghum cultivation areas. A total of 264 sorghum cultivation area images from Andong City, Gyeongsangbuk-do, were used for model training, and 14 images from Jecheon City, Chungcheongbuk-do, were used for testing. U-Net training was performed for 14 epochs, achieving AccuracyC = 0.9426, LossC = 0.1593, Dice_coefC = 0.9223, AccuracyV = 0.6403, LossV = 1.9624, Dice_coefV = 0.6402, with a training time of 4 hours and 37 minutes. Seg-Net training was performed for 101 epochs, achieving AccuracyC = 0.6363, LossC = 0.6573, Dice_coefC = 0.5586, AccuracyV = 0.5711, LossV = 0.6785, Dice_coefV = 0.5586, with a training time of 1 hour and 55 minutes. Test results showed that U-Net achieved AccuracyT = 0.6806, LossT = 0.7180, Dice_coefT = 0.5558, while Seg-Net achieved AccuracyT = 0.7472, LossT = 0.5225, Dice_coefC = 0.6159. Although Seg-Net showed lower calibration performance, it demonstrated superior generalization performance, faster training time, and lower memory requirements, making it a more efficient model for sorghum cultivation area segmentation than U-Net.
참고문헌 (25)
- Ashapure A, "Developing a machine learning based cotton yield estimation framework using multi-temporal UAS data", ISPRS J. of Photogrammetry and Remote Sensing, 169, pp.180~194 (2020)
- Bae H, "Quality characteristics of muffins added with whole waxy sorghum flour", KFCS, 28(4), pp.473~478 (2012)
- Buttar PK, "Satellite imagery analysis for crop type segmentation using U-net architecture", Procedia Computer Science, 235, pp.3418~3427 (2024)
- Chen R, "Predicting individual apple tree yield using UAV multi-source remote sensing data and ensemble learning", Computers and Electronics in Agriculture, 201, pp.107275~ (2022)
- Dahlberg J, "Sorghum : An alternative feed, hay and forage", Proceedings, 2015Western Alfalfa & Forage Symposium (2015)
- Delavarpour N, "A technical study on UAV characteristics for precision agriculture applications and associated practical challenges", Remote Sensing, 13(6), pp.1204~ (2021)
- "World food situation" (2023)[URL]
- García S, "A study of statistical techniques and performance measures for genetics-based machine learning : accuracy and interpretability", Soft Computing, 13, pp.959~977 (2009)
- García-Martínez H, "Digital count of corn plants using images taken by unmanned aerial vehicles and cross correlation of templates", Agronomy, 10(4), pp.469~ (2020)
- Hao Y, "SEG-Net: Deep learning grasping with a soft enveloping gripper", IEEE Robotics and Automation Letters, ~ (2024)
- Hasan M, "Deep learning precision farming: Tomato leaf disease detection by transfer learning", Proceedings of 2nd International Conference on Advanced Computing and Software Engineering (ICACSE) (2019)
- Hu J, "Pretrained deep learning networks and multispectral imagery enhance maize LCC, FVC, and maturity estimation", Remote Sensing, 16(5), pp.784~ (2024)
- Jayanthi M, "Comparative study of different techniques used for medical image segmentation of liver from abdominal CT scan", 2016 International Conference on Wireless Communications, Signal Processing and Networking (WiSPNET) (2016)
- Kazungu FK, "Overview of sorghum (Sorghum bicolor L.), its economic importance, ecological requirements and production constraints in Kenya", International J. of Plant & Soil Science, 35(1), pp.62~71 (2023)
- Lin Z, "Sorghum panicle detection and counting using unmanned aerial system images and deep learning", Frontiers in Plant Science, 11(5), pp.34853~ (2020)
- Marsalis M, "Dry matter yield and nutritive value of corn, forage sorghum, and BMR forage sorghum at different plant populations and nitrogen rates", Field Crops Research, 116(1-2), pp.52~57 (2010)
- Park KS, "Sorghum field segmentation with U-Net from UAV RGB", Korean J. of Remote Sensing, 39(5_1), pp.521~535 (2023)
- Park M, "Sorghum panicle detection using YOLOv5 based on RGB image acquired by UAV system", KSAFM, 24(4), pp.295~304 (2022)
- Prasad PV, "Growth and production of sorghum and millets", Soils, Plant Growth and Crop Production, 2 (2009)
- Ronneberger O, "Convolutional networks for biomedical image segmentation. Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015", 18th International Conference,Proceedings, Part III 18 (2015)
- Stehr NJ, "Drones : The newest technology for precision agriculture", Natural Sciences Education, 44(1), pp.89~91 (2015)
- Tejaswi VSD, "Computer-aided diagnosis of liver cancer with improved seg net and deep stacking ensemble model", Computational Biology and Chemistry, pp.108243~ (2024)
- Tsouros DC, "A review on UAV-based applications for precision agriculture", Information, 10(11), pp.349~ (2019)
- Yang S, "The application of unmanned aircraft systems to plant protection in China", Precision Agriculture, 19, pp.278~292 (2018)
- Zhang C, "The application of small unmanned aerial systems for precision agriculture : A review", Precision Agriculture, 13, pp.693~712 (2012)