TAPe technology - a new word in image recognition
Greetings, dear Gabrazhiteli!
Let me bring to your attention Comexp's blog, as well as TAPe technology, which allows you to very quickly recognize images even in conditions of strong interference.
Interesting? I ask for a cut
As you know, there are some standard methods of image processing that can cope with a fairly narrow class of technical problems. The tools used for image analysis can only work with minimal interference. If the interference becomes significant, then the reliability of image recognition is sharply reduced. At the same time, its processing time increases significantly, which negates all the tasks of detecting an object or tracking it. Still, the object had already disappeared long ago, and the system only half an hour later announced that it recognized it!
The classic image processing technology involves three main stages:
Prior to the development of TAPe technology, the first two stages did not have a theoretical solution, provided there was organized interference. This explains the lack of algorithms for reliable recognition of a person by his photograph, or television images in the control systems of various law enforcement agencies.
Maybe this is a fiction, and it is impossible to create such a mechanism? However, here, as always, nature itself challenges us, once again demonstrating its perfection. The human visual analyzer is so perfect that we can recognize an object not only in difficult weather conditions, but also when most of this object is hidden from us. If for us to identify, for example, a glass on one of its edges does not seem difficult, then for a machine this is an impossible task! While the machine works with a huge number of disconnected numbers, a person has a holistic perception, recognizing an object in large fragments by a small number of signs.
However, a solution to this problem was still found. Comexp invented and patented a technology that allows you to recognize an object even if a large part of it is blocked. This technology is called TAPe (Theory of Active Perception - Theory of Active Perception).
The absence of convolution made it possible to recognize the object in question much more efficiently than it did with pyramidal visual machines based on convolution. For example, pyramidal visual machines are 10 6 times more effective than matrix machines !
You can look at the technology in action on our website in the “Test” section, where various video clips are presented that are searched in real time from the weekly broadcasts of several television channels in different regions. We change the list of TV channels and regions for testing every week, the current list is always displayed in the “Test” section. I want to note that to reduce the load on the database, we turned on caching on the site, so a report for each fragment is generated for the first time after it is added.
You can also upload your video clip (registration in two steps is required) and see how quickly the database will be searched. If you don’t have any videos from the air at hand, then we have prepared a list of relevant commercials and screensavers for testing.
Commercials:
TV and TV screensavers:
I would like to know the opinion of the harazhiteli about our technology. Now we are developing the direction of monitoring television, as a priority in the use of TAPe technology, but in parallel we are developing APIs for third-party developers. Are there any ideas on this subject, wishes, suggestions? I will be glad to know your opinion on this issue.
Thanks for attention.
Let me bring to your attention Comexp's blog, as well as TAPe technology, which allows you to very quickly recognize images even in conditions of strong interference.
Interesting? I ask for a cut
Why is this needed?
As you know, there are some standard methods of image processing that can cope with a fairly narrow class of technical problems. The tools used for image analysis can only work with minimal interference. If the interference becomes significant, then the reliability of image recognition is sharply reduced. At the same time, its processing time increases significantly, which negates all the tasks of detecting an object or tracking it. Still, the object had already disappeared long ago, and the system only half an hour later announced that it recognized it!
The classic image processing technology involves three main stages:
- Primary processing
- Identification of classification features
- Image classification.
Prior to the development of TAPe technology, the first two stages did not have a theoretical solution, provided there was organized interference. This explains the lack of algorithms for reliable recognition of a person by his photograph, or television images in the control systems of various law enforcement agencies.
Maybe this is a fiction, and it is impossible to create such a mechanism? However, here, as always, nature itself challenges us, once again demonstrating its perfection. The human visual analyzer is so perfect that we can recognize an object not only in difficult weather conditions, but also when most of this object is hidden from us. If for us to identify, for example, a glass on one of its edges does not seem difficult, then for a machine this is an impossible task! While the machine works with a huge number of disconnected numbers, a person has a holistic perception, recognizing an object in large fragments by a small number of signs.
However, a solution to this problem was still found. Comexp invented and patented a technology that allows you to recognize an object even if a large part of it is blocked. This technology is called TAPe (Theory of Active Perception - Theory of Active Perception).
The stages of technology are as follows:
- Drawing up the initial description. Simultaneous processing of the entire image as a whole. The absence of a convolution procedure at this stage avoids dependence on possible interference (noise). The resulting interference becomes part of the image, which can be used in favor of the object in question: in some cases, this helps to obtain its informative elements; all existing pictures are identically displayed on a certain set of pictures, which can then be processed in real time using standard computers. For example, if we are talking about processing an image on a field of size 1024x1024 with gradation levels of 256, then the power of the variety of images (the whole set of images) is a cosmic number of 256 1048576. During processing, the power of many grayscale images is significantly reduced and can be represented by a small number of features that carry information about the most significant structural elements of the image.
- Image analysis. Processing the results of the first stage, spatial differentiation, which allows you to identify composite and structural relationships between all structural parts of the image.
- Creating an image of the image . Special graphic descriptions are prepared that reflect the skeletons and integral curves of the structural relationships of image elements.
- Making decisions. The compiled visual description is compared with the standard available in the database. If there is no standard, then a new standard is formed.
The absence of convolution made it possible to recognize the object in question much more efficiently than it did with pyramidal visual machines based on convolution. For example, pyramidal visual machines are 10 6 times more effective than matrix machines !
View in action
You can look at the technology in action on our website in the “Test” section, where various video clips are presented that are searched in real time from the weekly broadcasts of several television channels in different regions. We change the list of TV channels and regions for testing every week, the current list is always displayed in the “Test” section. I want to note that to reduce the load on the database, we turned on caching on the site, so a report for each fragment is generated for the first time after it is added.
You can also upload your video clip (registration in two steps is required) and see how quickly the database will be searched. If you don’t have any videos from the air at hand, then we have prepared a list of relevant commercials and screensavers for testing.
Commercials:
TV and TV screensavers:
I would like to know the opinion of the harazhiteli about our technology. Now we are developing the direction of monitoring television, as a priority in the use of TAPe technology, but in parallel we are developing APIs for third-party developers. Are there any ideas on this subject, wishes, suggestions? I will be glad to know your opinion on this issue.
Thanks for attention.
