Approach to Pattern Recognition

Pattern recognition is an inborn ability of animals, and pattern recognition includes recognizing faces, objects, words, etc. Pattern recognition is one of the basic core problems in cognitive psychology. Human pattern recognition can be viewed as a typical perception procedure that is based on the knowledge and experience that people have gained. Without using someone's knowledge and experience, a person cannot understand the value of the stimulating information pattern inserted, then neither can the pattern be recognized nor the objects.

What are Approaches to Pattern Recognition?

Pattern recognition is the basic human perception or intelligence which stands out prominently in different human activities. It is linked with different psychological processes such as sense, memory study, thinking, etc. pattern recognition is a very important way to get a cognitive view of human psychological activities. Visually pattern recognition is a process of inserting stimulating information, comparing them to the information of long-term memory, and then identifying the category to which stimulation belongs. Hence pattern recognition is dependent on an individual's knowledge and experience.

Principles of Pattern Recognition

Major principles are

• It should be done quickly and precisely
• It should be identified and then classified as unknown objects
• Accurately identify shapes and objects with the help of angles
• Patterns and objects should be identified even if they are partly concealed
• Identify patterns rapidly, with ease, and automatically

Theories of Pattern Recognition

There are mainly six theories of pattern recognition. They are

• Template Matching Theory of Pattern Recognition
• Prototype Matching Theory of Pattern Recognition
• Feature analysis theory of pattern recognition
• Recognition by components theory of pattern recognition
• Fourier analysis
• Bottom-up and top-down processing

Some of these approaches are discussed below.

Template Matching

Template matching is one of the easiest and earliest approaches to pattern recognition. The incoming sensory information is compared with stored templates, which are stored in the procedure of someone's experience and knowledge.

When a simulation acts upon people's sense organ, then at first, the simulating information is coded, compared, and then matched with the pattern stored in the brain, then recognized as a single particular pattern in the brain which become the best match. Some examples of template matching are common in our day-to-day life like the machine can quickly identify the seal on page X based on the comparison with the template. It also has some disadvantages. According to thiss theory, one must store a suitable template before identifying a template. Nevertheless, these templates are many, and it not only brings a heavy load to memory but also lead to less flexible and rigid pattern recognition.

Prototype Matching Theory

The prototype is referred to a concept of mean characteristics of a certain object. Unlike template matching, prototype matching does not highlight the best and most appropriate match between the incoming stimuli and the stored information in the brain. This theory tells us about the fundamental features of one type of object. According to the theory of prototype matching, in the procedure of identifying patterns, external simulations only need to be contrasted with the prototype, and sense to objects results from the matching between input facts and prototype. If the external simulating information matches appropriately with a particular prototype in the brain, then the information can be arranged or grouped in a particular prototype category and identified. For example, a small animal with feathers, a beak, and two wings that can fly is a prototype concept of a pigeon, crow, eagle, etc. Again, people have seen various kinds of airplanes, but the prototype of the airplane is a long cylinder with two wings. It is more economical than template theory and describes the speed of identification of letters, words, patterns, etc.

Feature Analysis Theory

According to this theory, the optical system breaks down the incoming stimuli into their characteristics and then processes the information. Here one tries to match the characteristics of the pattern with the stored information in memory, unlike the whole pattern with the template or the prototype. Some characteristics are maybe more valuable for identification than others. Each stimulus has a set of distinctive properties. This theory mainly states that our nervous systems have many receptors that filter various stimuli that reach our brains. The receptors are often termed feature detectors and can encode various features, facts, and details of a certain object. Feature analysis advances through four stages:

• Detection
• Pattern dissection
• Feature comparison in memory
• Recognition

This theory is currently the most famous model. For example, we have the stored fact that the alphabet Z comprises two horizontal lines and one oblique line with two acute angles. On the other hand, alphabet Y has two oblique lines, one vertical line, and one acute angle. When someone is shown a letter of the alphabet, the recognition procedure includes identifying the types of angles and lines and comparing them with the stored information of all alphabets. It is more flexible than template theory.

Top-Down and Bottom-Up Processing of Pattern Recognition

Here the stimulus information reaches the sensory receptors then the combination of bottom-level features permits the individual to identify more complex inter patterns. Our memory helps us identify patterns in how the world is organized. Bottom-up processing is data-driven processing, while top-down processing is concept-based processing.

Structural Theory of Pattern Recognition

This theory explains how features can join together to form a structure which displays a leading principle of Gestalt psychology. It mainly highlights the relations among the features. Gionee is defined as various three-dimensional shapes that combine to create three-dimensional patterns.

Conclusion

Thus, it can be concluded that pattern recognition happens when the information from the surrounding is received and entered into a short-term memory resulting in a natural activation of a particular content of long-term memory. Identifying patterns permits an individual to predict and expect what is coming in front of someone; hence, pattern recognition includes matching the facts received with the information already stored in the brain.