Öz
Savunma teknolojilerinde kullanılan termal optik sistemler, görüntüleme sırasında çeşitli sorunlarla karşılaşırlar ve bu sorunlar, görüntülerin üzerinde farklı türlerde bulanıklık kusuru olarak ortaya çıkarlar. Bulanıklığın türünü belirlemek, görüntüyü iyileştirmenin ilk adımıdır. Bu çalışmada, çeşitli bulanıklık türleri (odaksızlık bulanıklığı, atmosferik türbülans bulanıklığı, görüntü titreme bulanıklığı, Gaussian bulanıklığı ve hareket bulanıklığı) modellenmiş ve 15000 FLIR termal görüntü içeren bir veri seti üzerinde Python programlama dili aracılığıyla rastgele değerlerle bulanıklıklar uygulanarak bir veri seti oluşturulmuştur. Daha sonra, ResNet50, InceptionV3, DenseNet201, VGG16 ve EfficientNetB0 gibi farklı Evrişimsel Sinir Ağları bu termal görüntüleri sınıflandırmak için kullanılmıştır. Elde edilen sonuçlara göre en yüksek performans %98 doğruluk ile EfficientNetB0 mimarisi tarafından sağlanmıştır. Bu çalışma ile termal görüntülerdeki bulanıklık türlerini sınıflandırmada derin öğrenme yaklaşımının etkileri incelenmiş ve gelecekteki uygulamalar için umut verici sonuçlar elde edilmiştir.
Anahtar Kelimeler
Evrişimsel sinir ağları görüntü bulanıklığı görüntü sınıflandırma termal görüntüleme
Kaynakça
- Abramowitz, M., Davidson, MW. 2018. Troubleshooting microscope configuration and other common errors. Olympus Microscopy Resource Center, 1-14.
- Burduk, R. 2020. Classification performance metric for ımbalance data based on recall and selectivity normalized in class labels.
- Chong, RM., Tanaka, T. 2009. Image extrema analysis and blur detection with identification. 2008 IEEE International Conference on Signal Image Technology and Internet Based Systems, pp. 320 – 326, Bali, Indonesia. Doi: 10.1109/SITIS.2008.38
- Christen, P., Hand, D., Kirielle, N. 2023. A review of the F-Measure: Its History, properties, criticism, and alternatives. ACM Computing Surveys, 56(3):1-24. Doi: 10.1145/3606367.
- Fan, M., Huang, R., Feng, W., Sun, J. 2017. Image blur classification and blur usefulness assessment. 2017 IEEE International Conference on Multimedia & Expo Workshops (ICMEW), pp. 531-536, Hong Kong. Doi: 10.1109/ICMEW.2017.8026291
- FLIR ADAS Dataset v2.0.0. 2022. FREE Teledyne FLIR Thermal Dataset for algorithm training. https://www.flir.com/oem/adas/adas-dataset-form/
- Hussein, A., Garba Sharifai, A., Alia, O., Abualigah, L., Almotairi, K., Abujayyab, S., Gandomi, A. 2024. Auto-detection of the coronavirus disease by using deep convolutional neural networks and X-ray photographs. Scientific Reports, 14(1):534. Doi: 10.1038/s41598-023-47038-3
- Kılıç, İ., Yalçın, N. 2024. Transfer öğrenme teknikleri kullanarak nohut çeşidi sınıflandırma. Karaelmas Fen Ve Mühendislik Dergisi, 14(1):48-58. DOI: 10.7212/karaelmasfen.1427715
- Lawson, JK., Carrano, CJ. 2006. Using historic models of Cn2 to predict r0 and regimes affected by atmospheric turbulence for horizontal, slant, and topological paths. Atmospheric Optical Modeling, Measurement, and Simulation II, SPIE, 6303:38-49. Doi: 10.1117/12.679108
- Liu, R., Li, Z., Jia, J. 2008. Image partial blur detection and classification. IEEE International Conference on Computer Vision Pattern Recognition, pp. 1-8, Anchorage, AK. Doi: 10.1109/CVPR.2008.4587465
- Liu, S., Lin, F., Tong, X., Zhang, H., Lin, H., Xie, H., Ye, Z., Zheng, S. 2022. Dynamic PSF-based jitter compensation and quality improvement for push-broom optical images considering terrain relief and the TDI effect. Applied Optics, 61(16):4655-4662. Doi: 10.1364/AO.453163
- Pei, Y., Huang, Y., Zou, Q., Xy, Z., Wang, S. 2019. Effects of image degradation and degradation removal to CNN-based image classification. IEEE Transactions on Pattern Analysis and Machine Intelligence, 43(4):1239-1253. Doi: 10.1109/TPAMI.2019.2950923
- Sakthivel, S., Akash, K., Sidarth, B., Ganesh, C. 2022. Blur classification and estimation of motion blur parameters using OLR. International Conference on Electronic Systems and Intelligent Computing (ICESIC), pp. 323-328, Chennai, India. Doi: 10.1109/ICESIC53714.2022.9783566
- Tiwari, S. 2018. Blur classification using segmentation based fractal texture analysis. Indonesian Journal of Electrical Engineering and Informatics. 6(4):373-384. Doi: 10.11591/ijeei.v6i4.463
- Tiwari, S. 2020. A blur classification approach using deep convolution neural network. International Journal of Information System Modeling and Design. 11(1): 93-111. Doi: 10.4018/IJISMD.2020010106
- Wang, R., Li, W., Qin, R., Wu, J. 2017. Blur image classification based on deep learning. 2017 IEEE International Conference on Imaging Systems and Techniques (IST), pp. 1-6, Beijing, China. Doi: 10.1109/IST.2017.8261503
- Wu, J., Su, X. 2018. Method of ımage quality improvement for atmospheric turbulence degradation sequence based on graph Laplacian filter and nonrigid registration. Mathematical Problems in Engineering. 2018(1): 4970907. Doi: 10.1155/2018/4970907
- Zhang, K., Ren, W., Luo, W., Lai, W., Stenger, B., Yang, MH., Li, H. 2022. Deep image deblurring: A Survey. International Journal of Computer Vision. 130:1-28. Doi: 10.1007/s11263-022-01633-5
Öz
Thermal optical systems used in defense technologies encounter various challenges during imaging, which manifest as different types of blurring artifacts on the images. Identifying the type of blur represents the first step in enhancing the image clarity. In this study, various types of blurs (defocus blur, atmospheric turbulence blur, image jitter blur, Gaussian blur, and motion blur) are modeled in Python, and a blurred dataset is created by applying these blurs with random values to a dataset containing 15000 FLIR thermal images. Subsequently, different Convolutional Neural Networks architectures such as ResNet50, InceptionV3, DenseNet201, VGG16, and EfficientNetB0 are used to classify these thermal images. This study represents a significant step for future applications by evaluating the effects of the deep learning approach in classifying blur types in thermal images. According to the results obtained, the highest performance is achieved with an accuracy of 98% using the EfficientNetB0 architecture. This study examines the effects of the deep learning approach in classifying blur types in thermal images and yields promising results for future applications.
Anahtar Kelimeler
Convolutional neural networks image blur image classification thermal imaging
Kaynakça
- Abramowitz, M., Davidson, MW. 2018. Troubleshooting microscope configuration and other common errors. Olympus Microscopy Resource Center, 1-14.
- Burduk, R. 2020. Classification performance metric for ımbalance data based on recall and selectivity normalized in class labels.
- Chong, RM., Tanaka, T. 2009. Image extrema analysis and blur detection with identification. 2008 IEEE International Conference on Signal Image Technology and Internet Based Systems, pp. 320 – 326, Bali, Indonesia. Doi: 10.1109/SITIS.2008.38
- Christen, P., Hand, D., Kirielle, N. 2023. A review of the F-Measure: Its History, properties, criticism, and alternatives. ACM Computing Surveys, 56(3):1-24. Doi: 10.1145/3606367.
- Fan, M., Huang, R., Feng, W., Sun, J. 2017. Image blur classification and blur usefulness assessment. 2017 IEEE International Conference on Multimedia & Expo Workshops (ICMEW), pp. 531-536, Hong Kong. Doi: 10.1109/ICMEW.2017.8026291
- FLIR ADAS Dataset v2.0.0. 2022. FREE Teledyne FLIR Thermal Dataset for algorithm training. https://www.flir.com/oem/adas/adas-dataset-form/
- Hussein, A., Garba Sharifai, A., Alia, O., Abualigah, L., Almotairi, K., Abujayyab, S., Gandomi, A. 2024. Auto-detection of the coronavirus disease by using deep convolutional neural networks and X-ray photographs. Scientific Reports, 14(1):534. Doi: 10.1038/s41598-023-47038-3
- Kılıç, İ., Yalçın, N. 2024. Transfer öğrenme teknikleri kullanarak nohut çeşidi sınıflandırma. Karaelmas Fen Ve Mühendislik Dergisi, 14(1):48-58. DOI: 10.7212/karaelmasfen.1427715
- Lawson, JK., Carrano, CJ. 2006. Using historic models of Cn2 to predict r0 and regimes affected by atmospheric turbulence for horizontal, slant, and topological paths. Atmospheric Optical Modeling, Measurement, and Simulation II, SPIE, 6303:38-49. Doi: 10.1117/12.679108
- Liu, R., Li, Z., Jia, J. 2008. Image partial blur detection and classification. IEEE International Conference on Computer Vision Pattern Recognition, pp. 1-8, Anchorage, AK. Doi: 10.1109/CVPR.2008.4587465
- Liu, S., Lin, F., Tong, X., Zhang, H., Lin, H., Xie, H., Ye, Z., Zheng, S. 2022. Dynamic PSF-based jitter compensation and quality improvement for push-broom optical images considering terrain relief and the TDI effect. Applied Optics, 61(16):4655-4662. Doi: 10.1364/AO.453163
- Pei, Y., Huang, Y., Zou, Q., Xy, Z., Wang, S. 2019. Effects of image degradation and degradation removal to CNN-based image classification. IEEE Transactions on Pattern Analysis and Machine Intelligence, 43(4):1239-1253. Doi: 10.1109/TPAMI.2019.2950923
- Sakthivel, S., Akash, K., Sidarth, B., Ganesh, C. 2022. Blur classification and estimation of motion blur parameters using OLR. International Conference on Electronic Systems and Intelligent Computing (ICESIC), pp. 323-328, Chennai, India. Doi: 10.1109/ICESIC53714.2022.9783566
- Tiwari, S. 2018. Blur classification using segmentation based fractal texture analysis. Indonesian Journal of Electrical Engineering and Informatics. 6(4):373-384. Doi: 10.11591/ijeei.v6i4.463
- Tiwari, S. 2020. A blur classification approach using deep convolution neural network. International Journal of Information System Modeling and Design. 11(1): 93-111. Doi: 10.4018/IJISMD.2020010106
- Wang, R., Li, W., Qin, R., Wu, J. 2017. Blur image classification based on deep learning. 2017 IEEE International Conference on Imaging Systems and Techniques (IST), pp. 1-6, Beijing, China. Doi: 10.1109/IST.2017.8261503
- Wu, J., Su, X. 2018. Method of ımage quality improvement for atmospheric turbulence degradation sequence based on graph Laplacian filter and nonrigid registration. Mathematical Problems in Engineering. 2018(1): 4970907. Doi: 10.1155/2018/4970907
- Zhang, K., Ren, W., Luo, W., Lai, W., Stenger, B., Yang, MH., Li, H. 2022. Deep image deblurring: A Survey. International Journal of Computer Vision. 130:1-28. Doi: 10.1007/s11263-022-01633-5
Ayrıntılar
| Birincil Dil | Türkçe |
|---|---|
| Konular | Bilgisayar Yazılımı, Yazılım Mühendisliği (Diğer) |
| Bölüm | Araştırma Makaleleri |
| Yazarlar | |
| Yayımlanma Tarihi | 22 Nisan 2025 |
| Gönderilme Tarihi | 1 Temmuz 2024 |
| Kabul Tarihi | 15 Ekim 2024 |
| Yayımlandığı Sayı | Yıl 2025 Cilt: 15 Sayı: 1 |
