Bölütleme Tabanlı Yeni Görüntü İyileştirme Yöntemi

Year 2021, Issue: 32, 975 - 981, 31.12.2021

Nurullah Öztürk , Serkan Öztürk

https://doi.org/10.31590/ejosat.1041197

Abstract

Histogram eşitleme yöntemi, görüntüde kontrastı ve parlaklığı ayarlamak için kullanılan temel görüntü işleme yöntemidir. Ancak histogram eşitleme, görüntülerde aşırı iyileşme, yapaylık, doygunluk ve ayrıntıların kaybolması gibi olumsuzluklar oluşturabilmektedir. Bu çalışmada görüntü bölütleme tabanlı yeni görüntü iyileştirme yöntemi önerilmiştir. Önerilen yöntemde görüntüdeki nesne bölgeleri aktif kontur tabanlı yöntemler ile bölütlenmiş ve bu bölgelerde histogram eşitleme uygulanmıştır. Daha sonra elde edilen iyileştirilmiş nesneler, giriş görüntüsündeki bölgesine eklenmiştir. Önerilen bu yöntem ile histogram eşitleme yönteminin görüntüler üzerinde oluşturduğu olumsuz etkiler önlenerek daha etkili iyileştirmeler sağlanmıştır. Ayrıca bölütleme yöntemi, histogram genişletme ve bi-histogram eşitleme yöntemleri ile birleştirilerek mevcut yöntemlerin başarısı da incelenmiştir. Görüntülerin entropi değeri, mutlak ortalama parlaklık hatası (AMBE) ve Tepe-Sinyal-Gürültü-Oranı (PSNR) metrikleri performans karşılaştırmasında kullanılmıştır. Elde edilen sonuçlar görsel ve sayısal olarak verilmiştir. Önerilen yöntem, mevcut histogram eşitleme tabanlı yöntemler ile karşılaştırılmış ve yöntemin başarısı ortaya çıkarılmıştır.

Keywords

Görüntü iyileştirme, Görüntü bölütleme, Aktif kontur yöntemi, Histogram eşitleme

A New Image Enhancement Method Based on Segmentation

Abstract

The histogram equalization method is the fundamental image processing method used to adjust the contrast and brightness in the image. However, histogram equalization can cause negative effects such as excessive enhancement, artifacts, saturation, and loss of details in images. In this paper, a segmentation-based new image enhancement method is proposed. With this proposed method, more effective enhancement is obtained by preventing the negative effects of the histogram equalization method on images. In the proposed method, the object regions in the image are segmented with active contour-based methods, and histogram equalization is applied to these regions. Enhanced objects obtained later are added to their region in the input image. With this proposed method, more effective enhancement is achieved by preventing the negative effects of the histogram equalization method on images. In addition, the success of the existing methods is examined by combining the segmentation method with histogram stretching and bi-histogram equalization methods. The entropy value of the images, the absolute average luminance error (AMBE), and the Peak-Signal-Noise-Ratio (PSNR) metrics are used in the performance comparison. The obtained results are presented both visually and numerically. The proposed method is compared with the histogram equalization-based methods, and the success of the proposed method is revealed.

Keywords

Image enhancement, Image segmentation, Active contour method, Histogram equalization.

References

• Gonzalez, R. C. ve Woods, R. E. (2014). Digital Image Processing (Sayısal Görüntü İşleme), cilt 3, Pearson Education (Çeviri Palme).
• Alasu, S. (2018). Çizge Kesim Temelli İnteraktif Görüntü Bölütleme Yöntemlerinin Karşılaştırılması (Yüksek Lisans Tezi) İnönü Üniversitesi Fen Bilimleri Enstitüsü, Malatya.
• Bhargavi, K. ve Jyothi, S. (2014). A survey on threshold based segmentation technique in image processing. International Journal of Innovative Research and Development, 3(12), 234-239.
• Iannizzotto, G. ve Vita, L. (2000). Fast and accurate edge-based segmentation with no contour smoothing in 2-D real images. IEEE Transactions on Image Processing, 9(7), 1232-1237.
• Karthick, S., Sathiyasekar, K., ve Puraneeswari, A. (2014). A survey based on region based segmentation. International Journal of Engineering Trends and Technology, 7(3), 143-147.
• Zou, Y. ve Liu, B. (2016). Survey on clustering-based image segmentation techniques. In 2016 IEEE 20th International Conference on Computer Supported Cooperative Work in Design (CSCWD) (pp. 106-110). IEEE.
• Kass, M., Witkin, A., ve Terzopoulos, D. (1988). Snakes: Active contour models. International journal of computer vision, 1(4), 321-331.
• Gupta, P., Kumare, J. S., Singh, U. P., ve Singh, R. K. (2017). Histogram based image enhancement techniques: a survey. Int J Comput Sci Eng, 5(6), 475-484.
• Pizer, S. M., Amburn, E. P., Austin, J. D., Cromartie, R., Geselowitz, A., Greer, T., ... ve Zuiderveld, K. (1987). Adaptive histogram equalization and its variations. Computer vision, graphics, and image processing, 39(3), 355-368.
• Kim, Y. T. (1997). Contrast enhancement using brightness preserving bi-histogram equalization. IEEE transactions on Consumer Electronics, 43(1), 1-8.
• Wang, Y., Chen, Q., ve Zhang, B. (1999). Image enhancement based on equal area dualistic sub-image histogram equalization method. IEEE transactions on Consumer Electronics, 45(1), 68-75.
• Chen, S. D., ve Ramli, A. R. (2003). Minimum mean brightness error bi-histogram equalization in contrast enhancement. IEEE transactions on Consumer Electronics, 49(4), 1310-1319.
• Singh, K., ve Kapoor, R. (2014). Image enhancement via median-mean based sub-image-clipped histogram equalization. Optik, 125(17), 4646-4651.
• Singh, K., ve Kapoor, R. (2014). Image enhancement using exposure based sub image histogram equalization. Pattern Recognition Letters, 36, 10-14.
• Ye, Z., Mohamadian, H., ve Ye, Y. (2008). Gray level image processing using contrast enhancement and watershed segmentation with quantitative evaluation. In 2008 International Workshop on Content-Based Multimedia Indexing (pp. 470-475). IEEE.
• Hung, C. S., ve Ruan, S. J. (2014). An intelligent block segmentation based contrast enhancement for edge preservation. In 2014 IEEE 3rd Global Conference on Consumer Electronics (GCCE) (pp. 260-261). IEEE.
• Liang, Y. T., Zhang, M., Zhao, K. B., ve Li, Y. G. (2016, December). Haze image moving window threshold segmentation algorithm based on contrast enhancement. In 2016 10th International Conference on Software, Knowledge, Information Management & Applications (SKIMA) (pp. 357-363). IEEE.
• Yin, H., Lyu, G., Luo, X., ve Li, C. (2017). A segmentation-based adaptive image enhancement method inspired by the self-adjust features of HVS. International Journal of Machine Learning and Cybernetics, 8(6), 1895-1905.
• Sahu, P. K., ve Bhawnani, D. K. (2014). Thyroid segmentation and area measurement using active contour. International Journal of Engineering and Advanced Technology (IJEAT), 3(5).
• Chan, T. F., ve Vese, L. A. (2001). Active contours without edges. IEEE Transactions on image processing, 10(2), 266-277.
• Li, C., Kao, C. Y., Gore, J. C., ve Ding, Z. (2007). Implicit active contours driven by local binary fitting energy. In 2007 IEEE Conference on Computer Vision and Pattern Recognition (pp. 1-7). IEEE.
• Chen, S. D., ve Ramli, A. R. (2003). Minimum mean brightness error bi-histogram equalization in contrast enhancement. IEEE transactions on Consumer Electronics, 49(4), 1310-1319.
• Wang, Z., Bovik, A. C., Sheikh, H. R., ve Simoncelli, E. P. (2004). Image quality assessment: from error visibility to structural similarity. IEEE transactions on image processing, 13(4), 600-612.