{"id":932,"date":"2017-12-08T21:31:02","date_gmt":"2017-12-08T21:31:02","guid":{"rendered":"https:\/\/ryma.cinvestav.mx\/ravg\/?post_type=project&#038;p=932"},"modified":"2017-12-08T21:31:02","modified_gmt":"2017-12-08T21:31:02","slug":"color-adaptive-training-underwater-color-restoration","status":"publish","type":"project","link":"https:\/\/ryma.cinvestav.mx\/ravg\/project\/color-adaptive-training-underwater-color-restoration\/","title":{"rendered":"Color Adaptive Training for Underwater Color Restoration"},"content":{"rendered":"<p>For underwater robotics applications involving monitoring\u00a0and inspection tasks, it is important to capture quality\u00a0color images in real time.\u00a0In the last years, there has been a great number of research\u00a0that focus on the color correction or color restoration of\u00a0underwater images, which differs from the color enhancement\u00a0problem. The first attempts to recover the <em>\u201ctrue\u201d<\/em> color of the\u00a0scene while the second tries to enhance the color and other\u00a0cues but not necessarily recovering the original tones. Most\u00a0of the work reported in the literature in the color restoration\u00a0problem rely on approaches that take the image formation\u00a0process as its baseline. However, the fact that many of the\u00a0variables involved in this process are constantly changing\u00a0makes it hard to model, therefore some assumptions have to be\u00a0made. Other approaches consider the use of image processing\u00a0techniques that are applied directly on the values of the pixels\u00a0in the image. There are few work that use statistical approaches\u00a0that learn the correlations between a set of color depleted\u00a0and color image patches in order to recover the color in new\u00a0images. Furthermore, the use of suitable color space models\u00a0has demonstrated to be crucial in the color restoration process.<\/p>\n<p class=\"p1\">We have\u00a0proposed a statistically\u00a0learning method with an automatic selection of the training set for\u00a0restoring the color of underwater color-degraded video sequences. Our method uses a Markov\u00a0Random Field (MRF) model, which, as in any statistical\u00a0learning method, strongly depends on a training set. In our\u00a0case, to estimate the missing color in a video sequence, the\u00a0training set must capture, along the whole video sequence,\u00a0the correlations between a color degraded image and its\u00a0corresponding color. Since we could not directly obtain true color\u00a0training patches coming from the underwater scenes (<em>i.e.<\/em>,\u00a0either these scenes where already taken or we are not allowed\u00a0to use a source of light at any time due to habitat protection\u00a0reasons), we first recover the color by applying a multiple\u00a0color space analysis and processing stage in a previously\u00a0selected image frame of the video sequence to be used as\u00a0a seed (training set) in our MRF-BP model. The adaptive\u00a0training is thought in terms of the way an image frame is\u00a0selected at a given moment according to the occurrence of a\u00a0change in the distribution of the channel values compared to\u00a0previous distributions.<\/p>\n<p>Experimental results in real underwater\u00a0video sequences demonstrate that our approach is feasible, even\u00a0when visibility conditions are poor, as our method can recover\u00a0and discriminate between different colors in objects that may\u00a0seem similar to the human eye.<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter  wp-image-933\" src=\"https:\/\/ryma.cinvestav.mx\/ravg\/wp-content\/uploads\/sites\/19\/2017\/12\/Results_color_restoration.jpg\" alt=\"\" width=\"444\" height=\"481\" srcset=\"https:\/\/ryma.cinvestav.mx\/ravg\/wp-content\/uploads\/sites\/19\/2017\/12\/Results_color_restoration.jpg 484w, https:\/\/ryma.cinvestav.mx\/ravg\/wp-content\/uploads\/sites\/19\/2017\/12\/Results_color_restoration-277x300.jpg 277w\" sizes=\"auto, (max-width: 444px) 100vw, 444px\" \/><\/p>\n<p>Comparison between the our AWW method and the WW assumption. (a) underwater images; (b) output using WW-Assumption (taken from [1]); and our adapted WW-assumption (c).<\/p>\n<p><strong>References<\/strong><\/p>\n<p>[1] \u00a0G. Bianco, M. Muzzupappa, R. Bruno, F.and Garcia, and L. Neumann. &#8220;A\u00a0new color correction method for \u00a0 \u00a0 \u00a0 \u00a0 underwater imaging. The International\u00a0Archives of Photogrammetry, Remote Sensing and Spatial \u00a0 \u00a0Information\u00a0Sciences, 40(5):25, 2015.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>For underwater robotics applications involving monitoring\u00a0and inspection tasks, it is important to capture quality\u00a0color images in real time.\u00a0In the last years, there has been a great number of research\u00a0that focus on the color correction or color restoration of\u00a0underwater images, which differs from the color enhancement\u00a0problem. The first attempts to recover the \u201ctrue\u201d color of the\u00a0scene [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":934,"comment_status":"open","ping_status":"closed","template":"","meta":{"_et_pb_use_builder":"","_et_pb_old_content":"","_et_gb_content_width":"","footnotes":""},"project_category":[],"project_tag":[],"class_list":["post-932","project","type-project","status-publish","has-post-thumbnail","hentry"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.3 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Color Adaptive Training for Underwater Color Restoration - Robotics Active Vision Group<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/ryma.cinvestav.mx\/ravg\/project\/color-adaptive-training-underwater-color-restoration\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Color Adaptive Training for Underwater Color Restoration - Robotics Active Vision Group\" \/>\n<meta property=\"og:description\" content=\"For underwater robotics applications involving monitoring\u00a0and inspection tasks, it is important to capture quality\u00a0color images in real time.\u00a0In the last years, there has been a great number of research\u00a0that focus on the color correction or color restoration of\u00a0underwater images, which differs from the color enhancement\u00a0problem. The first attempts to recover the \u201ctrue\u201d color of the\u00a0scene [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/ryma.cinvestav.mx\/ravg\/project\/color-adaptive-training-underwater-color-restoration\/\" \/>\n<meta property=\"og:site_name\" content=\"Robotics Active Vision Group\" \/>\n<meta property=\"og:image\" content=\"https:\/\/ryma.cinvestav.mx\/ravg\/wp-content\/uploads\/sites\/19\/2017\/12\/Example_color_restoration.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"627\" \/>\n\t<meta property=\"og:image:height\" content=\"174\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data1\" content=\"2 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/project\\\/color-adaptive-training-underwater-color-restoration\\\/\",\"url\":\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/project\\\/color-adaptive-training-underwater-color-restoration\\\/\",\"name\":\"Color Adaptive Training for Underwater Color Restoration - Robotics Active Vision Group\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/project\\\/color-adaptive-training-underwater-color-restoration\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/project\\\/color-adaptive-training-underwater-color-restoration\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/wp-content\\\/uploads\\\/sites\\\/19\\\/2017\\\/12\\\/Example_color_restoration.jpg\",\"datePublished\":\"2017-12-08T21:31:02+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/project\\\/color-adaptive-training-underwater-color-restoration\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/project\\\/color-adaptive-training-underwater-color-restoration\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/project\\\/color-adaptive-training-underwater-color-restoration\\\/#primaryimage\",\"url\":\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/wp-content\\\/uploads\\\/sites\\\/19\\\/2017\\\/12\\\/Example_color_restoration.jpg\",\"contentUrl\":\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/wp-content\\\/uploads\\\/sites\\\/19\\\/2017\\\/12\\\/Example_color_restoration.jpg\",\"width\":627,\"height\":174},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/project\\\/color-adaptive-training-underwater-color-restoration\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Projects\",\"item\":\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/project\\\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"Color Adaptive Training for Underwater Color Restoration\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/#website\",\"url\":\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/\",\"name\":\"Robotics Active Vision Group\",\"description\":\"Miembro de Rob\u00f3tica y Manufactura Avanzada - Cinvestav\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/ryma.cinvestav.mx\\\/ravg\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Color Adaptive Training for Underwater Color Restoration - Robotics Active Vision Group","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/ryma.cinvestav.mx\/ravg\/project\/color-adaptive-training-underwater-color-restoration\/","og_locale":"en_US","og_type":"article","og_title":"Color Adaptive Training for Underwater Color Restoration - Robotics Active Vision Group","og_description":"For underwater robotics applications involving monitoring\u00a0and inspection tasks, it is important to capture quality\u00a0color images in real time.\u00a0In the last years, there has been a great number of research\u00a0that focus on the color correction or color restoration of\u00a0underwater images, which differs from the color enhancement\u00a0problem. The first attempts to recover the \u201ctrue\u201d color of the\u00a0scene [&hellip;]","og_url":"https:\/\/ryma.cinvestav.mx\/ravg\/project\/color-adaptive-training-underwater-color-restoration\/","og_site_name":"Robotics Active Vision Group","og_image":[{"width":627,"height":174,"url":"https:\/\/ryma.cinvestav.mx\/ravg\/wp-content\/uploads\/sites\/19\/2017\/12\/Example_color_restoration.jpg","type":"image\/jpeg"}],"twitter_card":"summary_large_image","twitter_misc":{"Est. reading time":"2 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/ryma.cinvestav.mx\/ravg\/project\/color-adaptive-training-underwater-color-restoration\/","url":"https:\/\/ryma.cinvestav.mx\/ravg\/project\/color-adaptive-training-underwater-color-restoration\/","name":"Color Adaptive Training for Underwater Color Restoration - Robotics Active Vision Group","isPartOf":{"@id":"https:\/\/ryma.cinvestav.mx\/ravg\/#website"},"primaryImageOfPage":{"@id":"https:\/\/ryma.cinvestav.mx\/ravg\/project\/color-adaptive-training-underwater-color-restoration\/#primaryimage"},"image":{"@id":"https:\/\/ryma.cinvestav.mx\/ravg\/project\/color-adaptive-training-underwater-color-restoration\/#primaryimage"},"thumbnailUrl":"https:\/\/ryma.cinvestav.mx\/ravg\/wp-content\/uploads\/sites\/19\/2017\/12\/Example_color_restoration.jpg","datePublished":"2017-12-08T21:31:02+00:00","breadcrumb":{"@id":"https:\/\/ryma.cinvestav.mx\/ravg\/project\/color-adaptive-training-underwater-color-restoration\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/ryma.cinvestav.mx\/ravg\/project\/color-adaptive-training-underwater-color-restoration\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/ryma.cinvestav.mx\/ravg\/project\/color-adaptive-training-underwater-color-restoration\/#primaryimage","url":"https:\/\/ryma.cinvestav.mx\/ravg\/wp-content\/uploads\/sites\/19\/2017\/12\/Example_color_restoration.jpg","contentUrl":"https:\/\/ryma.cinvestav.mx\/ravg\/wp-content\/uploads\/sites\/19\/2017\/12\/Example_color_restoration.jpg","width":627,"height":174},{"@type":"BreadcrumbList","@id":"https:\/\/ryma.cinvestav.mx\/ravg\/project\/color-adaptive-training-underwater-color-restoration\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/ryma.cinvestav.mx\/ravg\/"},{"@type":"ListItem","position":2,"name":"Projects","item":"https:\/\/ryma.cinvestav.mx\/ravg\/project\/"},{"@type":"ListItem","position":3,"name":"Color Adaptive Training for Underwater Color Restoration"}]},{"@type":"WebSite","@id":"https:\/\/ryma.cinvestav.mx\/ravg\/#website","url":"https:\/\/ryma.cinvestav.mx\/ravg\/","name":"Robotics Active Vision Group","description":"Miembro de Rob\u00f3tica y Manufactura Avanzada - Cinvestav","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/ryma.cinvestav.mx\/ravg\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"}]}},"_links":{"self":[{"href":"https:\/\/ryma.cinvestav.mx\/ravg\/wp-json\/wp\/v2\/project\/932","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/ryma.cinvestav.mx\/ravg\/wp-json\/wp\/v2\/project"}],"about":[{"href":"https:\/\/ryma.cinvestav.mx\/ravg\/wp-json\/wp\/v2\/types\/project"}],"author":[{"embeddable":true,"href":"https:\/\/ryma.cinvestav.mx\/ravg\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/ryma.cinvestav.mx\/ravg\/wp-json\/wp\/v2\/comments?post=932"}],"version-history":[{"count":1,"href":"https:\/\/ryma.cinvestav.mx\/ravg\/wp-json\/wp\/v2\/project\/932\/revisions"}],"predecessor-version":[{"id":935,"href":"https:\/\/ryma.cinvestav.mx\/ravg\/wp-json\/wp\/v2\/project\/932\/revisions\/935"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/ryma.cinvestav.mx\/ravg\/wp-json\/wp\/v2\/media\/934"}],"wp:attachment":[{"href":"https:\/\/ryma.cinvestav.mx\/ravg\/wp-json\/wp\/v2\/media?parent=932"}],"wp:term":[{"taxonomy":"project_category","embeddable":true,"href":"https:\/\/ryma.cinvestav.mx\/ravg\/wp-json\/wp\/v2\/project_category?post=932"},{"taxonomy":"project_tag","embeddable":true,"href":"https:\/\/ryma.cinvestav.mx\/ravg\/wp-json\/wp\/v2\/project_tag?post=932"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}