{"id":8612,"date":"2012-12-08T01:09:41","date_gmt":"2012-12-07T23:09:41","guid":{"rendered":"http:\/\/hgpu.org\/?p=8612"},"modified":"2012-12-08T01:09:41","modified_gmt":"2012-12-07T23:09:41","slug":"a-computationally-efficient-parallel-kernel-regression-for-image-reconstruction","status":"publish","type":"post","link":"https:\/\/hgpu.org\/?p=8612","title":{"rendered":"A Computationally Efficient Parallel Kernel Regression for Image Reconstruction"},"content":{"rendered":"<p>Image reconstruction is a method by which the underlying images, hidden in blurry and noisy data, can be retrieved. This is used in applications such as computer tomography (CT), magnetic resonance and radio astronomy. In recent times, a non-parametric adaptive regression method called steering kernel regression was proposed and proved to be effective. This method involves computation of local gradient at each pixel thereby making it computationally intensive. The time consuming parts of the steering kernel regression can be optimized by off-loading them onto GPUs and multi-core processors. The parallel implementation of this algorithm improved the performance of image processing applications such as denoising, deblocking and upscaling. It has given an average speedup factor of 6 on multi-core and 21 on a GPU.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Image reconstruction is a method by which the underlying images, hidden in blurry and noisy data, can be retrieved. This is used in applications such as computer tomography (CT), magnetic resonance and radio astronomy. In recent times, a non-parametric adaptive regression method called steering kernel regression was proposed and proved to be effective. This method [&hellip;]<\/p>\n","protected":false},"author":351,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":false,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[36,89,33,38,3],"tags":[1787,14,1786,512,1788,20,1274,567],"class_list":["post-8612","post","type-post","status-publish","format-standard","hentry","category-algorithms","category-nvidia-cuda","category-image-processing","category-medicine","category-paper","tag-algorithms","tag-cuda","tag-image-processing","tag-image-reconstruction","tag-medicine","tag-nvidia","tag-tesla-t20","tag-tomography"],"views":2687,"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/8612","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/users\/351"}],"replies":[{"embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=8612"}],"version-history":[{"count":0,"href":"https:\/\/hgpu.org\/index.php?rest_route=\/wp\/v2\/posts\/8612\/revisions"}],"wp:attachment":[{"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=8612"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=8612"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hgpu.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=8612"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}