A Morphological Filtering-Based Image-Enhancement Method for Citrus Plant Diseases
Bobbinpreet Kaur, Tripti Sharma, Bhawna Goyal, Ayush Dogra · 2021
The images captured though acquisition devices have some non-ideal characteristics such as low object-to-background contrast and high noise levels. Because of these characteristics it is necessary to correct the image and make it more informative. This can be achieved by applying image-enhancement techniques which work on statistical characteristics of an image to make it visually more pleasing and noise free. The main objective is to improve the image contrast and sharpen the edges to enhance the details in the image. In this article we propose a method using spatial and morphological filters. The spatial filter is applied in red, green and blue (RGB) decomposed image planes followed by a morphological filter. This approach improves the visual quality and information content in the image. The performance of the method is tested on different images with different variations of structuring elements. The considerable improvement in the values of peak signal-to-noise ratio (PSNR) is achieved by deployment of this hybrid filtering method.