Morphological contrast enhancement is proposed as an image enhancement technique. It involves adding bright areas and subtracting dark areas from the original image. This morphological contrast enhancement is applied in the pre-processing stage before image segmentation. Compared to histogram equalization, morphological contrast enhancement produces higher silhouette values, suggesting it can be used for gesture recognition. Future work may employ deep learning networks to reduce the number of steps needed for object recognition.
Morphological contrast enhancement is proposed as an image enhancement technique. It involves adding bright areas and subtracting dark areas from the original image. This morphological contrast enhancement is applied in the pre-processing stage before image segmentation. Compared to histogram equalization, morphological contrast enhancement produces higher silhouette values, suggesting it can be used for gesture recognition. Future work may employ deep learning networks to reduce the number of steps needed for object recognition.
Morphological contrast enhancement is proposed as an image enhancement technique. It involves adding bright areas and subtracting dark areas from the original image. This morphological contrast enhancement is applied in the pre-processing stage before image segmentation. Compared to histogram equalization, morphological contrast enhancement produces higher silhouette values, suggesting it can be used for gesture recognition. Future work may employ deep learning networks to reduce the number of steps needed for object recognition.
Morphological contrast enhancement is proposed as an image enhancement technique. It involves adding bright areas and subtracting dark areas from the original image. This morphological contrast enhancement is applied in the pre-processing stage before image segmentation. Compared to histogram equalization, morphological contrast enhancement produces higher silhouette values, suggesting it can be used for gesture recognition. Future work may employ deep learning networks to reduce the number of steps needed for object recognition.
Image Enhancement Using Morphological Contrast Enhancement
Abstract
In image processing, image enhancement plays an important role, as it is capable of
enhancing images that has been critically influenced by luminance or noise as well as repairing the damaged images to an extent. In this paper, image enhancement using morphological operations is proposed. The contrast enhancement process involves the addition of bright areas to original image and subtracting by dark areas. Morphological contrast enhancement is applied to the pre-processing stage to generate image that is given as the input of segmentation process. A comparison of the pre-processing process based on the quality of clusters produced through the segmentation process is detailed in this work. Silhouette coefficient is used as a measure to validate the accuracy of cluster generated in the segmentation process. The silhouette width value of 1, computed from the analysis of the segmentation, is considered as a good cluster. Furthermore, it can be observed from the result that the silhouette value obtained from the morphological contrast enhancement is higher than that provided by histogram equalization method. These scores suggest that the processed methods can be used in gesture recognition. As future scope deep network can be employed for gesture recognition reducing the number of processes required to obtain proper object recognition. References: [1] Munir Oudah, Ali-Al-Naji and Javaan Chahl, “Hand Gesture Recognition Based on Computer Vision: A Review of Techniques”, Journal of Imaging, July 2020. [2] I. M. O. Widyantara, I. M. D. P. Asana, N. M. A. E. D. Wirastuti and I. B. P. Adnyana, “Image enhancement using morphological contrast enhancement for video based image analysis,” 2016 International Conference on Data and Software Engineering (ICoDSE), Denpasar, 2016, pp. 1-6, doi: 10.1109/ICODSE.2016.7936115. [3] K. Sreedhar and B.Panlal, “Enhancement of images using morphological transformations”, International Journal of Computer Science & Information Technology (UCSIT) vol. 4, no. 1, February 2012.
Presented by, Guided by,
Anju Roslin Francis Dr. Geevarghese Titus S7 ECE A, Roll. No: 15 Associate Professor KTU Reg.No: AJC17EC016 Dept. of ECE, AJCE