One of the areas of machine learning that we do at Whiteaster is the recognition of 3D images and objects. How is this technology created and in which industries is image recognition used?
One of the areas of machine learning that we do at Whiteaster is the recognition of 3D images and objects. How is this technology created and in which industries is image recognition used?
The first stage is image acquisition. Then the image is analysed and the characteristics of the image are determined. This is a very important stage, because image description algorithms cannot (generally) struggle with the millions of pixels that make up the original image, but they need to get an interpretable representation of the image as a single point in the feature space.
The next step is to find a proper image description in the form of a properly selected mathematical formula. Finally, the proper recognition of the image and the related decision follows.
This is a classic approach to the subject of image classification recognition of patterns. Currently, the model of revolutionary networks is used, where the selection of essential features is done on its own at the stage of learning the network (the network learns the selection of features and their classification).
The systems are used, for example, in shops, using CCTV cameras to minimise the losses associated with minor thefts by recognising the faces of people previously involved in thefts. Similar systems are used at airports where the systems are able to detect the faces of people who, for various reasons, should not be on board.
The system would enable the pattern recognition of the building which the user takes a photo of by sending the server geolocation data from the built-in GPS receiver together with the photo taken. The server makes a list of objects which are in the vicinity and compares them with the sent photo. It then recognises the object and sends the information back to the device.
Mobile applications for taking pictures of products and automatically purchasing them in an online shop.
It is worth remembering that these are only some of the industries and sectors in which automatic image recognition technology is being developed. Artificial intelligence will soon dominate almost all areas of life.
Machines also learn about sounds (e.g. speech), illnesses (to support medical diagnosis), the economic situation (to optimise financial decisions), geological formations (to find new deposits of raw materials), or social moods (to predict the results of choices).