A face tracking picture optimization storage method in a security system
A technology for security system and optimized storage, which is applied in the field of optimized storage for face tracking images
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0080] Embodiments of the present invention provide a face tracking picture optimization storage method in a security system, such as figure 1 shown, including:
[0081] Step 1: Analyze the facial features of the captured historical face pictures, and store the historical face feature parameters in the database;
[0082] Step 2: Obtain the current face picture to be stored and the current face feature parameters, compare the current face feature parameters with the historical face feature parameters in the database, and obtain the historical face with the highest matching degree picture;
[0083] Step 3: Differentiate the current face picture and the historical face picture with the highest matching degree to obtain difference data, and store the difference data.
[0084] In this embodiment, the historical facial feature parameters include feature parameters such as eyes, nose, mouth, eyebrows, and cheeks.
[0085] The beneficial effects of the above design scheme are: by c...
Embodiment 2
[0087] On the basis of Embodiment 1, the present invention provides a method for optimizing and storing face tracking pictures in a security system. In step 1, the captured historical face pictures are subjected to face feature analysis, and the historical face feature parameters are stored. to the database include:
[0088] Carry out image enhancement and image standardization processing to the historical face images;
[0089] Input the processed history face picture into the artificial neural network model for training, obtain the first feature, calculate the error between the first feature and the standard face feature, and judge whether the error is within the preset range;
[0090] If so, using the first feature as the second feature;
[0091] Otherwise, adjust the number of hidden layers in the artificial neural network model, and perform retraining to obtain the second feature;
[0092] After the norm of the second feature is accumulated, the square root is obtained t...
Embodiment 3
[0097] On the basis of Embodiment 1, the present invention provides a face tracking picture optimization storage method in a security system. In step 2, obtaining the current face picture to be stored and the current face feature parameters include:
[0098] Capture the current face picture to be stored, and obtain the brightness detection value of the pixels of the current face picture;
[0099] Divide the luminance value in advance according to preset rules to generate several luminance value areas, respectively divide the pixels into the several luminance value areas based on the luminance detection value according to the preset rules, and obtain the several luminance value areas. The brightness value area includes the brightness average value of the brightness value area with the most pixels, and uses the brightness average value as the ambient brightness value of the surrounding environment of the captured current face picture to be stored;
[0100] judging whether the am...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More