Face recognition technology specifically refers to computer technology that uses analysis and comparison. Face recognition is a popular field of computer technology research, face tracking detection, automatic adjustment of image magnification, night-time infrared detection, and automatic adjustment of exposure intensity; it belongs to biometric recognition technology and is to the organism (generally to people ) Its own biological characteristics to distinguish between individual organisms.
The face recognition technology actually includes a series of related technologies for constructing a face recognition system, including face image collection, face location, face recognition preprocessing, identity verification, and identity search; and narrow face recognition specifically refers to people. The face of identity or identity search technology or system.
Face recognition technology principle:
Face recognition technology consists of three parts:
1, face detection <br> <br> face detection means determines a dynamic complex background scene and whether there is the image surface, and separating the image of such surface. There are generally the following methods:
1 The reference template method first designs one or several standard face templates, and then calculates the degree of matching between the samples collected by the test and the standard template, and determines whether there are human faces through the threshold.
2 Face Rule Method Because the face has a certain structural distribution characteristics, the so-called face rule method is to extract these features and generate corresponding rules to judge whether the test sample contains a face.
3 Sample learning method This method uses the method of artificial neural network in pattern recognition, that is, the classifier is generated by learning the face sample set and non-face sample set.
4 skin color model method This method is based on the appearance of skin color in the color space in the relative concentration of the law to detect.
5 Feature sub-face method This method treats all face image collections as a face sub-space and determines whether there is a face image based on the distance between the detected sample and its projection in the sub-space.
It is worth mentioning that the above five methods can also be comprehensively adopted in practical detection systems.
2, face tracking <br> <br> face tracking refers to the detected face for dynamic target tracking. Specifically, a model-based approach or a combination of motion and model approach is used. In addition, using skin color model tracking is also a simple and effective means.
3, or face verification target search image library in the face than <br> <br> face alignment is to be detected to look like identity. This actually means that the sampled face image is compared with the face image of the stock in order and the best match object is found. Therefore, the description of the face image determines the specific method and performance of face recognition. Two description methods are mainly used: feature vector and texture template:
â‘ eigenvector method This method is to determine the iris of the eye, nose, mouth and other surfaces like facial features of size, position, distance and other attributes, and then calculate their geometric feature amount, the feature amounts forming a description of the face image Feature vector.
2 face pattern template method This method is to store a number of standard face image templates or face image organ templates in the library. When performing the comparison, all the pixels of the sampled surface are used to measure the normalized correlation quantities of all the templates in the library. match. In addition, there is a method of using autocorrelation of pattern recognition or a combination of features and templates.
The core reality of face recognition technology is “local human character analysis†and “graphic/neural recognition algorithm.†This algorithm is a method that uses various organs and feature parts of the human face. For example, the corresponding data of multiple geometric identifications and the original parameters in the database are compared, judged and confirmed. The general requirement is to judge that the time is less than 1 second.
The face recognition process is generally divided into three steps:
1. First, create a face image file of a person's face. That is, using a camera to capture face images of a person's face or taking photos of them to form a face image file, and storing these facial image file face prints.
2. Get the current body image. That is, using the camera to capture the face image of the current person entering or exiting, or taking a photo input, and generating a face print code for the current face image file.
3. Compare the current profile code with archive inventory. Compare the face pattern of the current face image with the face line code in the file inventory. The above "face coding" method works based on the essential features and the beginning of the face. This face code is resistant to changes in light, skin tone, facial hair, hair style, eyeglasses, expressions, and gestures, and is highly reliable, enabling it to accurately identify a person from millions of people. The face recognition process can be accomplished automatically, continuously, and in real time using an ordinary image processing device.
The face recognition technology actually includes a series of related technologies for constructing a face recognition system, including face image collection, face location, face recognition preprocessing, identity verification, and identity search; and narrow face recognition specifically refers to people. The face of identity or identity search technology or system.
Face recognition technology principle:
Face recognition technology consists of three parts:
1, face detection <br> <br> face detection means determines a dynamic complex background scene and whether there is the image surface, and separating the image of such surface. There are generally the following methods:
1 The reference template method first designs one or several standard face templates, and then calculates the degree of matching between the samples collected by the test and the standard template, and determines whether there are human faces through the threshold.
2 Face Rule Method Because the face has a certain structural distribution characteristics, the so-called face rule method is to extract these features and generate corresponding rules to judge whether the test sample contains a face.
3 Sample learning method This method uses the method of artificial neural network in pattern recognition, that is, the classifier is generated by learning the face sample set and non-face sample set.
4 skin color model method This method is based on the appearance of skin color in the color space in the relative concentration of the law to detect.
5 Feature sub-face method This method treats all face image collections as a face sub-space and determines whether there is a face image based on the distance between the detected sample and its projection in the sub-space.
It is worth mentioning that the above five methods can also be comprehensively adopted in practical detection systems.
2, face tracking <br> <br> face tracking refers to the detected face for dynamic target tracking. Specifically, a model-based approach or a combination of motion and model approach is used. In addition, using skin color model tracking is also a simple and effective means.
3, or face verification target search image library in the face than <br> <br> face alignment is to be detected to look like identity. This actually means that the sampled face image is compared with the face image of the stock in order and the best match object is found. Therefore, the description of the face image determines the specific method and performance of face recognition. Two description methods are mainly used: feature vector and texture template:
â‘ eigenvector method This method is to determine the iris of the eye, nose, mouth and other surfaces like facial features of size, position, distance and other attributes, and then calculate their geometric feature amount, the feature amounts forming a description of the face image Feature vector.
2 face pattern template method This method is to store a number of standard face image templates or face image organ templates in the library. When performing the comparison, all the pixels of the sampled surface are used to measure the normalized correlation quantities of all the templates in the library. match. In addition, there is a method of using autocorrelation of pattern recognition or a combination of features and templates.
The core reality of face recognition technology is “local human character analysis†and “graphic/neural recognition algorithm.†This algorithm is a method that uses various organs and feature parts of the human face. For example, the corresponding data of multiple geometric identifications and the original parameters in the database are compared, judged and confirmed. The general requirement is to judge that the time is less than 1 second.
The face recognition process is generally divided into three steps:
1. First, create a face image file of a person's face. That is, using a camera to capture face images of a person's face or taking photos of them to form a face image file, and storing these facial image file face prints.
2. Get the current body image. That is, using the camera to capture the face image of the current person entering or exiting, or taking a photo input, and generating a face print code for the current face image file.
3. Compare the current profile code with archive inventory. Compare the face pattern of the current face image with the face line code in the file inventory. The above "face coding" method works based on the essential features and the beginning of the face. This face code is resistant to changes in light, skin tone, facial hair, hair style, eyeglasses, expressions, and gestures, and is highly reliable, enabling it to accurately identify a person from millions of people. The face recognition process can be accomplished automatically, continuously, and in real time using an ordinary image processing device.
Automatic Sizing Machine,Outer And Inner Ring Machine,Forge Oval Machine,Forge Oval Sizing Machine
Shaoxing Sunfong Intelligent Equipment Inc. , https://www.zjsxsunfong.com