[et_pb_section fb_built=”1″ admin_label=”section” _builder_version=”4.16″ background_image=”https:\/\/www.subsea-tech.com\/wp-content\/uploads\/2017\/12\/top_portiquedga.jpg” custom_padding=”240px||40px|” transparent_background=”off” global_colors_info=”{}”][et_pb_row make_equal=”on” admin_label=”row” _builder_version=”4.16″ background_size=”initial” background_position=”top_left” background_repeat=”repeat” global_colors_info=”{}”][et_pb_column type=”4_4″ _builder_version=”4.16″ custom_padding=”|||” custom_css_main_element=”display: flex; ||flex-direction: column; ||justify-content: center;” global_colors_info=”{}” custom_padding__hover=”|||”][et_pb_text admin_label=”Text” _builder_version=”4.18.0″ background_size=”initial” background_position=”top_left” background_repeat=”repeat” background_layout=”dark” hover_enabled=”0″ use_border_color=”off” border_style=”solid” global_colors_info=”{}” sticky_enabled=”0″ text_font_size=”18px”]<\/p>\n
Innovation\u00a0> Machines sp\u00e9ciales<\/p>\n
Client\/Utilisateur final : DGA<\/strong><\/p>\n [\/et_pb_text][\/et_pb_column][\/et_pb_row][\/et_pb_section][et_pb_section fb_built=”1″ _builder_version=”4.16″ transparent_background=”off” global_colors_info=”{}”][et_pb_row _builder_version=”4.16″ background_size=”initial” background_position=”top_left” background_repeat=”repeat” custom_padding=”27px|0px|0px|0px” global_colors_info=”{}”][et_pb_column type=”4_4″ _builder_version=”4.16″ custom_padding=”|||” global_colors_info=”{}” custom_padding__hover=”|||”][et_pb_text _builder_version=”4.18.0″ background_size=”initial” background_position=”top_left” background_repeat=”repeat” hover_enabled=”0″ global_colors_info=”{}” sticky_enabled=”0″]<\/p>\n Dans le cadre d’un programme de R&D li\u00e9 \u00e0 la d\u00e9tection de mines sous-marines, la DGA a confi\u00e9 \u00e0 SUBSEA TECH un projet dont le but \u00e9tait de d\u00e9terminer de mani\u00e8re aussi exhaustive que possible l’influence de divers param\u00e8tres de prises de vue sous-marines sur la qualit\u00e9 de l’image obtenue.<\/p>\n Un banc d’acquisition automatis\u00e9 a \u00e9t\u00e9 d\u00e9velopp\u00e9 pour cela. Il a permis de g\u00e9rer de mani\u00e8re rapide et pr\u00e9cise les tr\u00e8s nombreuses combinaisons de param\u00e8tres de prise de vue: position\/orientation des capteurs (cam\u00e9ras vid\u00e9os et sonars) et des sources lumineuses, types et nombre d’\u00e9clairages, turbidit\u00e9, type de fond (sable, rocher, algues\u2026), etc\u2026<\/p>\n Pour chaque image acquise, des techniques de segmentation originales ont \u00e9t\u00e9 d\u00e9velopp\u00e9es afin d’\u00e9valuer si l’information contenue dans l’image est exploitable ou pas en termes de d\u00e9tection et d’identification de cibles, et de constituer au final une base de donn\u00e9es \u00e0 l’aide de techniques pouvant s’apparenter \u00e0 du “big learning”.<\/p>\n [\/et_pb_text][et_pb_gallery gallery_ids=”5819,5821,5817,5815″ gallery_captions=”,,,” show_title_and_caption=”off” show_pagination=”off” _builder_version=”4.16″ custom_css_main_element=”||” global_colors_info=”{}”][\/et_pb_gallery][\/et_pb_column][\/et_pb_row][\/et_pb_section][et_pb_section fb_built=”1″ _builder_version=”4.16″ background_color=”#f9f9f9″ transparent_background=”off” global_colors_info=”{}”][et_pb_row _builder_version=”4.16″ background_size=”initial” background_position=”top_left” background_repeat=”repeat” global_colors_info=”{}” column_structure=”1_2,1_4,1_4″][et_pb_column type=”1_2″ _builder_version=”4.16″ custom_padding=”|||” global_colors_info=”{}” custom_padding__hover=”|||”][et_pb_text _builder_version=”4.18.0″ background_size=”initial” background_position=”top_left” background_repeat=”repeat” hover_enabled=”0″ global_colors_info=”{}” sticky_enabled=”0″]<\/p>\n Des techniques originales de traitement d’images vid\u00e9o et sonar sous-marines (notamment segmentation) et de gestion de base de donn\u00e9es, ont \u00e9t\u00e9 d\u00e9velopp\u00e9es pour ce projet, et associ\u00e9es \u00e0 des algorithmes d’auto-apprentissage.<\/p>\n [\/et_pb_text][\/et_pb_column][et_pb_column type=”1_4″ _builder_version=”4.16″ custom_padding=”|||” global_colors_info=”{}” custom_padding__hover=”|||”][et_pb_text _builder_version=”4.18.0″ background_size=”initial” background_position=”top_left” background_repeat=”repeat” hover_enabled=”0″ global_colors_info=”{}” sticky_enabled=”0″]<\/p>\n [\/et_pb_text][\/et_pb_column][et_pb_column type=”1_4″ _builder_version=”4.16″ custom_padding=”|||” global_colors_info=”{}” custom_padding__hover=”|||”][et_pb_text _builder_version=”4.18.0″ background_size=”initial” background_position=”top_left” background_repeat=”repeat” hover_enabled=”0″ global_colors_info=”{}” sticky_enabled=”0″]<\/p>\n [\/et_pb_text][\/et_pb_column][\/et_pb_row][\/et_pb_section]<\/p>\n","protected":false},"excerpt":{"rendered":" Innovation\u00a0> Machines sp\u00e9ciales Image Acquisition DGA Banc d’acquisition automatis\u00e9e d’imagerie sous-marine Client\/Utilisateur final : DGAObjectifs Dans le cadre d’un programme de R&D li\u00e9 \u00e0 la d\u00e9tection de mines sous-marines, la […]<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"_et_pb_use_builder":"on","_et_pb_old_content":"","_et_gb_content_width":"","footnotes":""},"class_list":["post-5823","page","type-page","status-publish","hentry"],"yoast_head":"\nObjectifs<\/h2>\n
Particularit\u00e9s du d\u00e9veloppement<\/strong><\/h2>\n
Nos comp\u00e9tences<\/strong><\/h2>\n
\n
Chiffres cl\u00e9s<\/strong><\/h2>\n
\n