{"id":8904,"date":"2024-03-14T14:59:23","date_gmt":"2024-03-14T13:59:23","guid":{"rendered":"https:\/\/v2.vision-tools.com\/?p=8904"},"modified":"2024-07-15T10:09:38","modified_gmt":"2024-07-15T08:09:38","slug":"evaluations-with-artificial-intelligence","status":"publish","type":"post","link":"https:\/\/www.vision-tools.com\/en\/projects\/inspection\/evaluations-with-artificial-intelligence.html","title":{"rendered":"Evaluations with artificial intelligence"},"content":{"rendered":"\n
\n
\"banner\"\/<\/figure>\n\n\n\n
<\/div>\n\n\n\n

Deep Learning Vision Systems<\/h1>\n\n\n\n
\n
\n
\n

Examples of applications with the use of AI<\/h4>\n\n\n\n
\"Evaluation<\/a><\/figure>\n\n\n\n

The surface inspection of a curved housing is one of the most problematic tasks in conventional machine vision. When used with AI, a wide variety of NOK examples are labelled, taught and reliably found in the evaluation.<\/p>\n\n\n\n

\"Evaluation<\/a><\/figure>\n\n\n\n

Evaluation result of the determined housing defects. <\/p>\n\n\n\n

\"Example<\/a><\/figure>\n\n\n\n
\"Example<\/a><\/figure>\n\n\n\n

Character recognition, even of embossed characters cropped in the image – 99% recognition.<\/p>\n\n\n\n

\"Example<\/a><\/figure>\n\n\n\n

Type recognition based on a taught-in feature and completeness check of sealing rings – 99% recognition.<\/p>\n<\/div>\n<\/div>\n\n\n\n

\n
\n

AI technology for evaluating image data uses neural networks to achieve a learned knowledge that allows it to distinguish between anomalies, shapes and characters while tolerating natural deviations. Thus, artificial intelligence combines the superior flexibility of humans with the performance of a machine system.<\/p>\n\n\n\n

<\/div>\n\n\n\n

How is an AI machine vision system created?<\/strong><\/strong><\/h2>\n\n\n\n

To prepare a Deep Learning system for use to reliably evaluate images, it initially needs a sufficiently large and easily identifiable training set of images with which the artificial intelligence can acquire its knowledge.<\/p>\n\n\n\n

To learn the features (classification) that are important for the inspection task, so-called labels<\/strong> (rectangular bounding boxes) are needed, which are generated either automatically or manually in the image. Labelling tools help the user to create the classes and position the labels.<\/p>\n\n\n\n

The current state of hardware allows Deep Learning algorithms<\/strong> to handle a mass of data from high-resolution images to easily handle standard image processing applications such as classification, object detection and segmentation. For all areas, there are object recognition pre-trained networks<\/strong> that help to shorten the training time of a well-functioning model. A trained model then corresponds to the learned knowledge that can be drawn on during later evaluation by the AI.<\/p>\n\n\n\n

<\/div>\n\n\n\n

How many images does a training set need to contain?<\/strong><\/strong><\/h2>\n\n\n\n

This depends on many factors, such as the complexity of the inspection task, the amount of possible deviations and the size of the image data. For a simple inspection task, it may be sufficient to train with 25-50 images. However, as the complexity of the task increases, this may well extend to 500-1000 images and more. A variance of the images can also be created simply by artificially manipulating the image data, e.g. by twisting, distorting or changing the brightness.<\/p>\n\n\n\n

<\/div>\n\n\n\n
\n

Areas of application for Deep Learning<\/strong><\/h2>\n\n\n\n