Deep analysis of the introduction, principle, process and application prospect of image recognition

Image recognition technology is an important technology in the information age. Its purpose is to let computers replace humans to process large amounts of physical information. With the development of computer technology, human understanding of image recognition technology has become more and more profound. The process of image recognition technology is divided into information acquisition, preprocessing, feature extraction and selection, classifier design and classification decision. The introduction of image recognition technology, its technical principle and pattern recognition are briefly analyzed. Then the image recognition technology of neural network and the image recognition technology of nonlinear dimensionality reduction and the application of image recognition technology are introduced. It can be concluded that the application of image processing technology is extensive, and human life will not be able to leave the image recognition technology. It is of great significance to study image recognition technology.

1. Introduction of image recognition technology

Image recognition is an important area of ​​artificial intelligence. The development of image recognition has gone through three stages: text recognition, digital image processing and recognition, and object recognition. Image recognition, as the name implies, is to make various processing and analysis on the image, and finally identify the target we want to study. The image recognition referred to today is not only identified by the human eye but by computer technology. Although human recognition ability is very powerful, for the fast-growing society, human's own recognition ability can not meet our needs, so computer-based image recognition technology has been produced. It is like human beings studying biological cells. It is unrealistic to observe cells completely by the naked eye, which naturally produces instruments such as microscopes for accurate observation. Usually, when a field has a demand that cannot be solved by inherent technology, a corresponding new technology will be generated. The same is true for image recognition technology. This technology is created to allow computers to process large amounts of physical information instead of humans, and to solve information that is unrecognizable or has a very low recognition rate.

1.1 Principles of Image Recognition Technology

In fact, the principle behind image recognition technology is not difficult, but the information it has to deal with is rather cumbersome. Any processing technology of the computer is not produced out of thin air. It is inspired by the scholars to use the program to simulate it. There is no essential difference between the computer image recognition technology and the human image recognition, but the machine lacks the influence of human beings on the sense of difference and visual difference. Human image recognition is not only recognized by the memory of the entire image stored in the mind. We recognize images by relying on the characteristics of the images and classifying the images first, and then through the characteristics of each category. The image was identified, but many times we didn't realize it. When we see an image, our brain quickly senses whether we have seen this picture or a picture similar to it. In fact, in the middle of "seeing" and "sensing", a rapid identification process has been experienced. The process of identification is somewhat similar to the search. In the process, our brain recognizes the categories that have been classified in the memory to see if there is a memory that has the same or similar characteristics as the image, thereby identifying whether the image has been seen. The same is true for the image recognition technology of the machine, which discriminates the image by sorting and extracting important features and excluding redundant information. These features extracted by the machine are sometimes very noticeable and sometimes very common, which affects the rate of machine recognition to a large extent. In summary, in the visual recognition of a computer, the content of an image is usually described by image features.

1.2 Pattern recognition

Pattern recognition is an important part of artificial intelligence and information science. Pattern recognition refers to the process of analyzing and processing different forms of information representing things or phenomena to obtain a description, identification and classification of things or phenomena.

Computer image recognition technology is a process of simulating human image recognition. Pattern recognition is essential in the process of image recognition. Pattern recognition was originally a basic intelligence of human beings. However, with the development of computers and the rise of artificial intelligence, human pattern recognition has not been able to meet the needs of life, so humans hope to use computers to replace or expand part of human brain work. This pattern recognition of the computer is created. Simply put, pattern recognition is the classification of data. It is a science that is closely integrated with mathematics. Most of the ideas used are probability and statistics. Pattern recognition is mainly divided into three types: statistical pattern recognition, syntax pattern recognition, and fuzzy pattern recognition.

2. The process of image recognition technology

Since computer image recognition technology is the same as human image recognition, their processes are similar. The process of image recognition technology is divided into the following steps: information acquisition, preprocessing, feature extraction and selection, classifier design and classification decision.

The acquisition of information refers to the conversion of information such as light or sound into electrical information through sensors. That is to obtain the basic information of the research object and transform it into information that the machine can recognize by some means.

Preprocessing mainly refers to operations such as denoising, smoothing, and transforming in image processing, thereby enhancing important features of the image.

Feature extraction and selection means that in pattern recognition, feature extraction and selection are required. The simple understanding is that the images we study are various. If we want to distinguish them by some method, we must identify them by the characteristics of these images. The process of acquiring these features is feature extraction. The features obtained in feature extraction may not be useful for this recognition. At this time, useful features are extracted, which is the choice of features. Feature extraction and selection is one of the most critical techniques in the image recognition process, so the understanding of this step is the focus of image recognition.

The classifier design refers to a recognition rule through training, through which a feature classification can be obtained, so that the image recognition technology can obtain a high recognition rate. Classification decision-making refers to classifying the identified objects in the feature space to better identify which category the object under study belongs to.

3. Analysis of image recognition technology

With the rapid development of computer technology and the continuous advancement of technology, image recognition technology has been applied in many fields. On February 15, 2015, Sina Technology released a news report: "Microsoft recently published a research paper on image recognition. In a benchmark for image recognition, computer system recognition capabilities have surpassed humans. Humans are in the classification database. The image recognition error rate in Image Net is 5.1%, and the deep learning system of Microsoft Research Group can achieve an error rate of 4.94%." From this news we can see that image recognition technology has surpassed image recognition. The trend of human image recognition capabilities. This also shows that the future image recognition technology has greater research significance and potential. Moreover, computers do have advantages that humans cannot surpass in many respects, and it is precisely because of this that image recognition technology can bring more applications to human society.

3.1 Neural Network Image Recognition Technology

Neural network image recognition technology is a relatively new image recognition technology. It is an image recognition method based on traditional image recognition methods and neural network algorithm. The neural network here refers to the artificial neural network, which means that the neural network is not the real neural network of the animal itself, but is artificially generated by humans after imitating the animal neural network. In neural network image recognition technology, the neural network image recognition model that combines genetic algorithm with BP network is very classic and has its application in many fields. In the image recognition system, the neural network system is generally used to extract the features of the image, and then the features of the image are mapped to the neural network for image recognition and classification. Take the car photo automatic identification technology as an example. When the car passes, the detection device of the car itself will be inductive. At this point, the inspection device will enable the image acquisition device to capture the image of the front and back of the car. Once the image is acquired, the image must be uploaded to a computer for storage for identification. The final license plate location module extracts the license plate information, identifies the characters on the license plate and displays the final result. A template-based matching algorithm and an artificial neural network-based algorithm are used in the process of recognizing characters on the license plate.

3.2 Nonlinear dimensionality reduction image recognition technology

Computer image recognition technology is an abnormally high-dimensional recognition technology. Regardless of the resolution of the image itself, the data it produces is often multi-dimensional, which poses a very large problem for the identification of computers. The most direct and effective way to make a computer's ability to recognize efficiently is to reduce dimensionality. Dimensionality reduction is divided into linear dimensionality reduction and nonlinear dimensionality reduction. For example, principal component analysis (PCA) and linear singular analysis (LDA) are common linear dimensionality reduction methods, which are characterized by simplicity and ease of understanding. But through linear dimensionality reduction, the overall data set is processed, and the optimal low-dimensional projection of the entire data set is obtained. It is verified that this linear dimensionality reduction strategy has high computational complexity and occupies relatively more time and space. Therefore, an image recognition technology based on nonlinear dimensionality reduction is generated, which is an extremely effective nonlinear feature extraction method. . This technique can find the nonlinear structure of the image and can reduce it on the basis of not destroying its intrinsic structure, so that the image recognition of the computer is performed in the lowest possible dimension, thus increasing the recognition rate. For example, the number of dimensions required for a face image recognition system is usually very high, and its high complexity is undoubtedly a huge "disaster" for a computer. Due to the uneven distribution of face images in high-dimensional space, humans can obtain compact face images through nonlinear dimensionality reduction techniques, thereby improving the efficiency of face recognition technology.

3.3 Application and prospects of image recognition technology

Computer image recognition technology has applications in many fields such as public safety, biology, industry, agriculture, transportation, and medical. For example, license plate recognition systems for transportation; face recognition technology for public safety, fingerprint recognition technology; seed identification technology for agriculture, food quality detection technology; medical ECG recognition technology. With the continuous development of computer technology, image recognition technology is constantly being optimized, and its algorithms are constantly improving. Images are the main source of human access and exchange of information, so image-related image recognition technology must also be the focus of future research. In the future, computer image recognition technology is likely to emerge in more fields. Its application prospects are also limitless, and human life will be more inseparable from image recognition technology.

Although image recognition technology is just emerging technology, its application has been quite extensive. Moreover, image recognition technology is also constantly growing. With the continuous advancement of technology, human understanding of image recognition technology will be more profound. Future image recognition technologies will be more powerful and more intelligent in our lives, bringing significant applications to more areas of human society. In the age of informationization in the 21st century, we can't imagine what our life would be like after we left image recognition technology. Image recognition technology is an indispensable technology for human life now and in the future.