Image processing and identification of objects in vision systems

НамДУ илмий ахборотномаси - Научный вестник НамГУ 2020 йил 3-сон
228 
Figure 1 - Functional diagram of STV 
The basic methods of image processing: formation, segmentation, description and 
analysis are presented in the block diagram of processing (Figure 2). 
Figure 2 - The block diagram of image processing 
 
Software implementation of the proposed methods 
The developed methods and algorithms are implemented in the form of completed 
independent modules for their use in computer vision systems, for example, in remote 

---

НамДУ илмий ахборотномаси - Научный вестник НамГУ 2020 йил 3-сон
229 
sensing data processing systems, mobile robots control, and integrated circuit design. In 
addition, the developed software products can be the basis for the construction of 
independent systems for various purposes [3, 4]. 
The development of such systems is carried out according to the technology of 
object-oriented analysis of information processing processes. The functional structure of 
such processes is represented by a directed graph, at the vertices of which data processing 
processors are located, and arcs provide data transport between them. The software 
architecture (SA) of the STV is depicted in Figure 3. The classes present in Figure 3 
correspond in their purpose to the blocks of the functional diagram (Figure 1). Figure 3 
shows two types of relationships between objects (instances of system classes).
Figure 3 - STV software architecture 
A line with a dot indicates the association (semantic connection). It does not explain 
how classes communicate with each other (one can only note semantic dependence, 
indicating which roles classes play for each other). The use is shown by a simple line and 
means what the association turns into if it turns out that one of its parties (the client) uses 
the services of the other (the server). An example of client-server relationships in our 
model: a class object Display Manager (responsible for displaying information during 
processing) uses the data provided by the class object Pre-analyzer (corresponding to the 
preprocessing block), Image Storage (image memory) and Image Analyzer (secondary 
processing image analysis). This class architecture is general and implies further 
refinement at the micro level of software development. 
Below are a number of developments carried out in this direction. 
Automated system for processing digital photos of the layers of metallization of the crystal of 
the integrated circuit. 
The system allows: to obtain a complete image of an object from fragments, to 
perform its preliminary processing, analysis of the spectral histogram and segmentation; 
approximate the boundaries of the selected areas in straight line segments; to carry out 
identification and correction on the video image of the metallization of a given layer and 
inter-level contacts [4]. 
The system is an integral part of the operational analysis of electronic products, 
both own and competing enterprises. It allows you to facilitate the input of project 

---