Artificial vision and deep learning

Icono Visión artificial y deep learning

In Inndeo we develop projects of design and installation of systems of artificial vision for the industry, implementing software applications for the processing of images through advanced algorithms, tailored to each application according to the client. If required, we also incorporate Deep Learning algorithms for the automatic classification of images according to different parameters. The purpose of machine learning techniques is to independently differentiate the patterns, following artificial intelligence techniques of the supervised and unsupervised type. Both techniques are used in the classification and segmentation of images, being able to automate the parameterization of certain characteristics of the products and quality failures.

Visión artificial y deep learning

Chemical image

icono imagen químicaMultispectral imaging technology and its subsequent conversion to chemical imaging has been implemented in satellites and aeronautical platforms over the last three decades. The new spectral vision systems have proven to be very useful for detecting the chemical surface composition of certain chemical products, including through their packaging. In the field of food inspection, obtaining spectral data has a long tradition as a non-destructive tool for the study of surface chemical composition. The advances in the spectral vision equipment, the image processing equipment, the chemical image composition software and the deep learning techniques allow us today to perform the classification of images in real time of the process. At Inndeo we develop chemical image inspection applications for the industry, automating the processes of quality inspection and product classification.
Imagen química

Magnetic resonance image

Icono Resonancia magnética

We develop applications of food inspection by magnetic resonance in real time. These techniques, unlike X-rays, are completely harmless for food. We have experience in the analysis of meat products, capturing tomographies of the interior of food in temporary periods of the order of 20 ms. Later the images are processed and classified according to the quality control parameters required, applying artificial vision algorithms and Deep Learning techniques. This technology is applicable to any type of food product, and unlike superficial image techniques, they allow us to penetrate the interior. We analyzed the composition of the tissues (meat, fat and water), as well as their distribution, classifying the products according to various quality parameters and we were able to detect the internal contamination by small metallic particles (hypodermic needles and metal shavings).

Resonancia magnética

Development of artificial vision equipment

Artificial vision is having exponential growth within industrial applications, allowing decision making based on image classification.In Inndeo we develop applications using the different existing vision technologies: 3D vision, linear, matrix, hyperspectral, infrared, high speed, etc. These technologies allow reducing costs and increasing quality standards by automating decision making in industrial processes.

In Inndeo we develop customized vision projects with different technologies depending on the type of project. Inndeo also develops its own machine vision equipment for inspection, solving the classification and quality control needs demanded by the industry. This range of solutions is marketed under the Inspectra brand. You can obtain more information about these devices by clicking on the link on the website.

We collaborate with the leading companies in the market in camera manufacturing and distribution of processing software. Ask us and we will advise you on the best possible solutions to your need.