Security machine manufacturers explain five problems of face recognition system

With the continuous development of technology, our security requirements for daily life and social safety have also become higher, and gradually we have also come into contact with face recognition systems. Of course, the birth of a new product will also follow A series of problems appear, let's understand the following with Tianyan security inspection machine manufacturers.

First, the lighting problem

Illumination change is a key factor that affects the performance of face recognition, and the degree of solution to this problem is related to the success or failure of the practical process of face recognition. Due to the 3D structure of the human face, the shadows cast by the light will strengthen or weaken the original facial features. Especially at night, facial shadows caused by insufficient light can cause a sharp drop in recognition rate, making it difficult for the system to meet practical requirements. At the same time, theories and experiments also prove that the difference caused by different illumination of the same individual is greater than the difference between different individuals under the same illumination. The problem of lighting is an old problem in machine vision, especially in face recognition. The solutions to the lighting problem include three-dimensional image face recognition and thermal imaging face recognition. But these two technologies are far from mature, and the recognition effect is not satisfactory.

Second, the attitude problem

Face recognition is mainly based on human facial features. How to recognize facial changes caused by posture has become one of the difficulties of this technology. The pose problem involves changes in the face caused by the rotation of the head about three axes in a three-dimensional vertical coordinate system, where the depth of rotation in two directions perpendicular to the image plane will cause partial loss of facial information. This makes the pose problem a technical problem for face recognition.

There are relatively few studies on posture. At present, most face recognition algorithms mainly focus on front and quasi-front face images. When pitching or left and right sides are more powerful, the recognition rate of face recognition algorithms will also be A sharp decline.

Three, expression problems

Changes in expressions such as crying, laughing, and anger with large facial extent also reflect the accuracy of facial recognition. Existing technology handles these aspects quite well. Whether it is to open mouth or make some exaggerated expressions, the computer can correct it through three-dimensional modeling and posture expression correction.

Fourth, the occlusion problem

The occlusion problem is a very serious problem for face image collection under non-cooperative situations. Especially in the monitoring environment, the monitored object often wears glasses, hats and other accessories, which makes the collected face image may be incomplete, which affects the subsequent feature extraction and recognition, and even leads to face detection algorithms Of failure.

V. Age changes

With the change of age, a person may change from a teenager to a young man to an old man, and his appearance may change considerably, resulting in a decline in recognition rate. For different ages, the recognition rate of face recognition algorithms is also different. A direct example of this problem is the identification of identification photographs. The validity period in China is generally 20 years. During these 20 years, everyone's appearance will inevitably change considerably, so there are also great problems in identification.