Multi-area feature acquisition method of automotive components, identification method of automotive components and its application in smart glasses
A technology of identification method and collection method, applied in the field of automobile data identification, can solve the problems of indistinguishable items of different models, poor real-time processing ability, etc., to facilitate category search and brand positioning, improve speed, and reduce mutual waiting time.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0027] A method for collecting multi-regional features of automotive components. During the classification training of automotive components, the features used by the automotive component classifier are evenly divided into intervals; when identifying automotive components, the multi-regional features of the automotive components to be collected are parallel collection.
[0028] Wherein: the specific steps of carrying out interval uniform division of the features used by the automotive component classifier are: during the automotive component classification training, using different dimension feature descriptions for different types of automotive components; offline learning and obtaining the classifier. The dimensional features used to distinguish the automotive components of the classification training, and the frequency statistics of the classifier using the dimensional features; according to the statistical results, the used dimensional features are evenly divided into inter...
Embodiment 2
[0033] It has the same technical solution as that of Embodiment 1, and more specifically: a method for identifying automobile components, which uses the above-mentioned method for collecting multi-region features of automobile components to collect multi-region features for automobile components. The hash tables of Num1 automotive component classifiers are stored in the detection terminal, and the multi-region feature collection data of automotive components is identified by looking up the corresponding hash table, and the corresponding information of the automotive components is provided. If the multi-area feature collection data of an automobile component cannot determine the type of the automobile component by looking up the corresponding hash table, the hash table corresponding to the automobile component does not exist in the hash table of Num1 automobile component classifiers stored in the detection terminal, Send the multi-region feature collection data of the automotive...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


