ASTIS - PIC Systems AG

Together with the company Semantic System we develop solutions based on artificial intelligence for the field forensics. As always if one invests in a new, potentially revolutionary technology the question of where to begin arose. Great risks faced the possible chances. We were always conscious, however, that the possibilities are almost unlimited in the field of forensic technology

The advantages of the new technology for science

The secret of the neural code (genetic algorithm GA)

The neural code refers to the thought process in the brain. Until now science made an effort for understanding this process analyzing the brain reactions and the given neural structure and then noting down the stimulated communication. We believe that this method will not lead to a complete understanding of the brain functions. In practice the brain structure does not represent any locally defined information-register. Similar or identical information contents in different brains can be represented in totally different neural structures. These structures can change with every intake of data and can also be reorganized with already known data. For this reason it is absolutely necessary that we understand the neural code (also known as the genetic algorithm).

In order to understand the underlying mathematics of the thought process, we took another approach. The secret is to find and understand the genetic algorithm (GA). This algorithm does not only describe how and why the cells (Neurons) react and fire off so-called stimuli-histograms, but it also interprets their importance. Like this it is possible to distinguish the meaning of an ambiguous word like "bank" (either savings bank or the bank of a river ect.) from the context of the sentence or paragraphs.

The build of a biologically compatible neural network

The holosemantic braiding is the key to a compatible neural net. However there is more to consider than simply constructing neurons (cells) which are identical to the biological ideal. The innate ability to be self-organizing and self-learning is critical. The holosemantic braiding is not really and actually intelligent it only represents the storage and using of the information. The real intelligence arises in the interaction between the stimulation through the data (Inquiry, input through the user etc.) to the communication (the dialog) with the holosemantic braiding.

The thought process facilitates intelligence in the data processing

The combination of the holosemantic braiding with the genetic algorithm (GA) allows the stimulation of the biological thought process. In addition to this fantasy, expected or unexpected future can be produced. Through the extrapolation of the thought process it is possible to simulate future conclusions and developments. With our new processor science and trade can now copy the action of billions of biological cells and create a model of a biological brain. Intelligence can only be facilitated it is no given static structure or ability.

The first automatic solution to match soles

Recognize where the similarities are from which shoe make these tracks might derive and can they come from the same original sole (shoe put in custody)?

Original shoe vs reference track vs scene track

A system working automatically comparing and analyzing various (similar) shoe tracks in order to find out which trackscould derive from which shoe type (Brand, type, name etc.). In this case the system compares the tracks of various cases or of different criminal-technical registrations of suspects among each other as well as against reference track collections that are already available.
ASTIS is supplied together with an image data bank and image editing software as well as standardized repro installation for the pictures. This database and application, PIC IMAGE ACCESS, can offer the automatic comparisons of shoe tracks as well as manual processing and analysis functions. If the Matcher has found a result the expert can do further spontaneous analyses with the process camera.

The semiotic sample analysis

That is the combination of pure picture comparison (optical) with the "knowledge" of sole tracks, meaning the knowledge of the concept what traces of soles imply and how such traces should be interpreted. In this context semiotics means that the system can make the connection between the reality of the track its origin and its importance. Because only if this knowledge is advected into viewing and picture comparison can a human being or a system make conclusions from the track and its origin.