An Approach for Reducing Outliers of Non Local Means Image Denoising Filter

We propose an adaptive approach for non local means (NLM) image filtering termed as non local adaptive clipped means (NLACM), which reduces the effect of outliers and improves the denoising quality as compared to traditional NLM. Common method to neglect outliers from a data population is computation of mean in a range defined by mean and standard deviation. In NLACM we perform the median within the defined range based on statistical estimation of the neighbourhood region of a pixel to be denoised. As parameters of the range are independent of any additional input and is based on local intensity values, hence the approach is adaptive. Experimental results for NLACM show better estimation of true intensity from noisy neighbourhood observation as compared to NLM at high noise levels. We have verified the technique for speckle noise reduction and we have tested it on ultrasound (US) image of lumbar spine. These ultrasound images act as guidance for injection therapy for treatment of lumbar radiculopathy. We believe that the proposed approach for image denoising is first of its kind and its efficiency can be well justified as it shows better performance in image restoration.

Cobb Angle Measurement of Scoliosis with Reduced Variability

Cobb angle, which is a measure of spinal curvature is the standard method for quantifying the magnitude of Scoliosis related to spinal deformity in orthopedics. Determining the Cobb angle through manual process is subject to human errors. In this work, we propose a methodology to measure the magnitude of Cobb angle, which appreciably reduces the variability related to its measurement compared to the related works. The proposed methodology is facilitated by using a suitable new improved version of Non-Local Means for image denoisation and Otsus automatic threshold selection for Canny edge detection. We have selected NLM for preprocessing of the image as it is one of the fine states of art for image denoisation and helps in retaining the image quality. Trimmedmean, median are more robust to outliners than mean and following this concept we observed that NLM denoising quality performance can be enhanced by using Euclidean trimmed-mean replacing the mean. To prove the better performance of the Non-Local Euclidean Trimmed-mean denoising filter, we have provided some comparative study results of the proposed denoising technique with traditional NLM and NonLocal Euclidean Medians. The experimental results for Cobb angle measurement over intra observer and inter observer experimental data reveals the better performance and superiority of the proposed approach compared to the related works. MATLAB2009b image processing toolbox was used for the purpose of simulation and verification of the proposed methodology.