{"id":68,"date":"2020-11-08T20:21:17","date_gmt":"2020-11-08T20:21:17","guid":{"rendered":"http:\/\/photoboiler.com\/?page_id=68"},"modified":"2021-01-07T12:04:44","modified_gmt":"2021-01-07T12:04:44","slug":"teknologiat","status":"publish","type":"page","link":"https:\/\/www.photoboiler.com\/en\/teknologiat\/","title":{"rendered":"Computer Vision Toolbox"},"content":{"rendered":"

Computer vision applications in simple forms have been in use for decades. However, it was not until the 2010s that with deep neural networks the ability of computers to recognize images reached human levels. <\/p>\n\n\n\n

Computer vision takes the automatic control of processes to a whole other level. With the help of machine learning, systems can copy the intuition of even an experienced person, for example in quality monitoring. Systematically measured quality, on the other hand, enables accurate process optimization and rapid response to deviations.<\/p>\n\n\n\n

Classification<\/h2>\n\n\n\n

Classification was the first application where deep neural networks were successfully used. As the name implies, classification labels an image as belonging to a category. For example, the mushroom in the picture below is classified as russula paludosa. <\/p>\n\n\n\n

\"\"<\/figure><\/div>\n\n\n\n

The deep neural net searches the image for colors and contrasts and textures and forms in the deeper layers.<\/p>\n\n\n\n

For industrial use the classification is often a preprocessing stage where image is labeled as valid for further analysis. For example making sure the conveyor belt has material.<\/p>\n\n\n\n

\"\"<\/figure><\/div>\n\n\n\n

Object detection<\/h2>\n\n\n\n

In object detection, an area is found in the image where the desired object can be found. As an example in the figure below, the propeller has been identified and its surface area measured from the overall view.<\/p>\n\n\n\n

\"\"<\/figure><\/div>\n\n\n\n

This is particularly useful when dealing with large amounts of images and deciding if an image is worth saving for later inspection. <\/p>\n\n\n\n

Segmentation<\/h2>\n\n\n\n

Area segmentation or semantic segmentation is the most practical tool in industrial image analysis and computer vision. <\/p>\n\n\n\n

In semantic segmentation every pixel in the image is classified as belonging to one or more categories. This can be then used for example to determine the amount of good quality product, defects or background in the image. The numerical value is needed for automation, which still runs primarily on numeric values.<\/p>\n\n\n\n

Try the slider below to visualize the amount of bark left.<\/p>\n\n\n\n

\r\n\t\t
\r\n\t\t\t\"\"
\"\"<\/div><\/div>
\"\"<\/div>\r\n\t\t\t\t\t\t\t
<\/div>
<\/div>
<\/div><\/div>
<\/div><\/div>\r\n\t\t\t\t\t\t<\/div>\t\t\t\t\t<\/div>\r\n\t\t<\/div>\n<\/div><\/div>\n\n\n\n

Segmentation makes numeric production monitoring and optimization possible.<\/p>\n\n\n\n

Check out the applications <\/a>or contact-info<\/a> .<\/p>","protected":false},"excerpt":{"rendered":"

Konen\u00e4k\u00f6sovelluksia on yksinkertaisissa muodoissa ollut jo vuosikymmeni\u00e4 k\u00e4yt\u00f6ss\u00e4. Kuitenkin vasta 2010-luvulla syv\u00e4t neuroverkot r\u00e4j\u00e4yttiv\u00e4t potin ja tietokoneiden kyky tunnistaa kuvia saavutti ihmistason. Konen\u00e4k\u00f6 nostaa prosessien automaattisen valvonnan aivan toiselle tasolle. Koneoppimisen avulla j\u00e4rjestelm\u00e4t voivat kopioida kokeneenkin ihmisen intuition esimerkiksi laadun tarkkailussa. J\u00e4rjestelm\u00e4llisesti mitattu laatu taas mahdollistaa prosessin tarkan optimoinnin ja nopean reagoinnin poikkeamiin. Luokittelu Luokittelu…<\/p>\n

Read More<\/a><\/p>","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"_links":{"self":[{"href":"https:\/\/www.photoboiler.com\/en\/wp-json\/wp\/v2\/pages\/68"}],"collection":[{"href":"https:\/\/www.photoboiler.com\/en\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/www.photoboiler.com\/en\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/www.photoboiler.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.photoboiler.com\/en\/wp-json\/wp\/v2\/comments?post=68"}],"version-history":[{"count":31,"href":"https:\/\/www.photoboiler.com\/en\/wp-json\/wp\/v2\/pages\/68\/revisions"}],"predecessor-version":[{"id":220,"href":"https:\/\/www.photoboiler.com\/en\/wp-json\/wp\/v2\/pages\/68\/revisions\/220"}],"wp:attachment":[{"href":"https:\/\/www.photoboiler.com\/en\/wp-json\/wp\/v2\/media?parent=68"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}