Residual Network Architecture Model for Image Weather Classification
Main Article Content
Abstract
Weather classification plays an important role in many fields, including agriculture, transportation, and meteorology. Traditional methods for weather recognition are usually based on human observation or sensor networks, which are prone to errors and quite costly. To overcome the limitation, this research implements the Convolutional Neural Network method with a Residual Network model architecture for image-based weather classification. Using a dataset of 1,500 images categorized into five weather conditions cloudy, foggy, rainy, sunny and sunrise. The model training accuracy reached a level of 92%, while the validation accuracy reached a level of 94% and resulted in a testing accuracy of 86.7%. The model training accuracy was high for sunny and sunrise conditions. Accuracy was lower in rainy and foggy weather conditions. This research shows that the ResNet model architecture can provide a low-cost, efficient, and high-accuracy solution for weather classification.Weather classification plays an important role in many fields, including agriculture, transportation, and meteorology. Traditional methods for weather recognition are usually based on human observation or sensor networks, which are prone to errors and quite costly. To overcome the limitation, this research implements the Convolutional Neural Network method with a Residual Network model architecture for image-based weather classification. Using a dataset of 1,500 images categorized into five weather conditions cloudy, foggy, rainy, sunny and sunrise. The model training accuracy reached a level of 92%, while the validation accuracy reached a level of 94% and resulted in a testing accuracy of 86.7%. The model training accuracy was high for sunny and sunrise conditions. Accuracy was lower in rainy and foggy weather conditions. This research shows that the ResNet model architecture can provide a low-cost, efficient, and high-accuracy solution for weather classification.
Downloads
References
C. Lu, D. Lin, J. Jia, and C.-K. Tang, “Two-Class Weather Classification,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Jun. 2014, doi: 10.1109/cvpr.2014.475.
B. Zhao, X. Li, X. Lu, and Z. Wang, “A CNN–RNN architecture for multi-label weather recognition,” Neurocomputing, vol. 322, pp. 47–57, Sep. 2018, doi: 10.1016/j.neucom.2018.09.048.
K. K. Singh, A. K. Baxla, P. Singh, and P. K. Singh, “Weather based information on risk management in agriculture,” Clim. Chang. Agric. India Impact Adapt., pp. 206–216, 2018, doi: 10.1007/978-3-319-90086-5_16.
H. Xiao, F. Zhang, Z. Shen, K. Wu, and J. Zhang, “Classification of Weather Phenomenon From Images by Using Deep Convolutional Neural Network,” Earth Sp. Sci., vol. 8, no. 5, pp. 1–9, 2021, doi: 10.1029/2020EA001604.
J. Xia, D. Xuan, L. Tan, and L. Xing, “ResNet15: Weather Recognition on Traffic Road with Deep Convolutional Neural Network,” Adv. Meteorol., vol. 2020, 2020, doi: 10.1155/2020/6972826.
T. Yu, Q. Kuang, J. Hu, J. Zheng, and X. Li, “Global-Similarity Local-Salience Network for Traffic Weather Recognition,” pp. 4607–4615, 2021, doi: 10.1109/ACCESS.2020.3048116.
K. B. Jayanthi, “An efficient weather recognition algorithm on highway roads for vehicle guidance,” vol. 6, no. June, pp. 1212–1223, 2022.
J. Carlos, V. Guerra, Z. Khanam, S. Ehsan, R. Stolkin, and K. Mcdonald-maier, “Weather Classification: A new multi-class dataset, data augmentation approach and comprehensive evaluations of Convolutional Neural Networks.”
S. Chen, T. Shu, H. Zhao, and Y. Yan, “For Real-Time Multi-Label Weather Recognition,” no. Dl, pp. 1–18, 2023.
E. Ship, “Real-Time Weather Image Classification with SVM: A Feature-Based Approach.”
S. A. Rahayu, S. B. Sipayung, A. Witono, and L. S. Suprihatin, “The Performance of K-Nearest Neighbor (KNN) Approach for Estimating Extreme Rain Events based on CCTV Images Camera Data,” J. EECCIS (Electrics, Electron. Commun. Control. Informatics, Syst., vol. 17, no. 2, pp. 59–65, 2023, doi: 10.21776/jeeccis.v17i2.1649.
M. N. Khan and M. M. Ahmed, “Snow Detection using In-Vehicle Video Camera with Texture-Based Image Features Utilizing K-Nearest Neighbor, Support Vector Machine, and Random Forest,” Transp. Res. Rec., vol. 2673, no. 8, pp. 221–232, 2019, doi: 10.1177/0361198119842105.
O. K. Clustering et al., “Deep Learning (CNN) and Transfer Learning: A Review,” 2022, doi: 10.1088/1742-6596/2273/1/012029.
K. Audhkhasi, O. Osoba, and B. Kosko, “Noise-enhanced convolutional neural networks,” Neural Networks, vol. 78, pp. 15–13, 2016.
Y. Shi, Y. Li, J. Liu, X. Liu, and Y. L. Murphey, “Weather Recognition Based on Edge Deterioration and Convolutional Neural Networks,” Proc. - Int. Conf. Pattern Recognit., vol. 2018-Augus, pp. 2438–2443, 2018, doi: 10.1109/ICPR.2018.8546085.
I. Kharisudin, M. F. Az-Zahra, E. R. Winarti, and S. B. Waluya, “Deep convolutional neural networks for the detection of macular diseases from optical coherence tomography images,” J. Phys. Conf. Ser., vol. 1567, no. 2, 2020, doi: 10.1088/1742-6596/1567/2/022076.
K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” 3rd Int. Conf. Learn. Represent. ICLR 2015 - Conf. Track Proc., pp. 1–14, 2015.
M. Ferguson, R. Ak, Y. T. T. Lee, and K. H. Law, “Automatic localization of casting defects with convolutional neural networks,” Proc. - 2017 IEEE Int. Conf. Big Data, Big Data 2017, pp. 1726–1735, 2017, doi: 10.1109/BigData.2017.8258115.
M. Bansal, M. Kumar, M. Sachdeva, and A. Mittal, “Transfer learning for image classification using VGG19: Caltech-101 image data set,” J. Ambient Intell. Humaniz. Comput., vol. 14, no. 4, pp. 3609–3620, 2023, doi: 10.1007/s12652-021-03488-z.
G. Huang, Z. Liu, and K. Q. Weinberger, “Densely Connected Convolutional Networks,” Comput. Vis. Pattern Recognit., pp. 2261–2269, 2017.
Y. Wang and Y. Li, “Research on Multi-class Weather Classification Algorithm Based on Multi-model Fusion,” no. Itnec, pp. 2251–2255, 2020.
T. Zhou, X. Ye, H. Lu, X. Zheng, S. Qiu, and Y. Liu, “Dense Convolutional Network and Its Application in Medical Image Analysis,” Biomed Res. Int., vol. 2022, 2022, doi: 10.1155/2022/2384830.
K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2016-Decem, pp. 770–778, 2016, doi: 10.1109/CVPR.2016.90.
X. Yang, T. Xie, L. Liu, and D. Zhou, “Image super-resolution reconstruction based on improved Dirac residual network,” Multidimens. Syst. Signal Process., vol. 32, no. 4, pp. 1065–1082, 2021, doi: 10.1007/s11045-021-00773-0.
D. Mishkin, N. Sergievskiy, and J. Matas, “Systematic evaluation of convolution neural network advances on the Imagenet,” Comput. Vis. Image Underst., vol. 161, pp. 11–19, 2017, doi: 10.1016/j.cviu.2017.05.007.
B. Zhao, L. Hua, X. Li, X. Lu, and Z. Wang, “Weather recognition via classification labels and weather-cue maps,” Pattern Recognit., vol. 95, pp. 272–284, 2019, doi: 10.1016/j.patcog.2019.06.017.
M. R. Ibrahim, J. Haworth, and T. Cheng, “Weathernet: Recognising weather and visual conditions from street-level images using deep residual learning,” ISPRS Int. J. Geo-Information, vol. 8, no. 12, 2019, doi: 10.3390/ijgi8120549.
Article Sidebar
