Image de-noising is a process of eliminating the unwanted information occupied on the image during image acquisition and image digitization. The noise removal in an image is a challenging pre-process, concern to preserve minute details in the images. The selection of proper noise removal filter will make the sense of retaining the fine details in the image will helpful for feature extraction. Especially in the domain of sperm cell image processing, the better denoising technique is obligatory to maintain the fine details in the image. Generally these images are not clear due to different staining procedures followed during image acquisition. The various emerging noise removal filters are developed for microscopic medical images but they should be identified specifically with respect to the noise in the input image. Most of the microscopic medical images are degraded by Gaussian noise. So that the Gaussian noise is added artificially on the human spermatozoa cell image and de-noised by using several spatial filters. In this paper the performance of various de-noising filters is measured using objective fidelity criteria such as Mean Square Error (MSE), Peak Signal-to-Noise Ratio (PSNR) and Entropy. This performance analysis moves forward to have further research in denoising for microscopic sperm cell images.
展开▼
