Metallographic structure identification method, system and equipment and storage medium
A technology of metallographic structure and identification method, which is applied in the field of computer vision measurement metallographic detection, can solve problems such as unbalanced metallographic image samples, improve recognition accuracy, improve recognition accuracy and generalization ability, and false detection rate Reduced effect
Pending Publication Date: 2021-11-26
XI AN JIAOTONG UNIV
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] Aiming at the problem of unbalanced metallographic image samples, the object of the present invention is to provide a metallographic structure recognition method, system, equipment and storage medium, which improves the metallographic structure recognition rate of a few types of samples
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0061] A metallographic structure identification method based on computer vision is used to identify and judge the metallographic structure of heat-resistant steel for pressure-bearing devices. The specific steps are as follows:
[0062] In general, in the metallographic structure of heat-resistant steel for pressure devices, there are six common metallographic structures, namely ferrite, pearlite, bainite, reticular ferrite, Widmanstatten structure and martensitic structure. body. Among them, Widmanstatten structure and martensite are abnormal phases in heat-resistant steel for pressure-bearing devices, reticular ferrite, Widmanstatten structure and martensite are relatively rare, and the distribution of samples is uneven, and the data distribution of the six types of structures is uneven. Equilibrium, for the convenience of description, in this embodiment, ferrite, pearlite and bainite are defined as the majority category, and the remaining three phases are defined as the mi...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a metallographic structure identification method, system and device and a storage medium, the method comprises steps of collecting metal material metallographic images, forming an original data set, dividing a training set and a test set, and using a first deep learning model obtained under training of all samples for identifying most types of metallographic structures; and on the basis, training the model by using a few types of samples, and selecting samples difficult to recognize according to an output result of the model. And forming a new data set by the difficult-to-recognize samples and the minority-class samples for training to obtain a second deep learning model, and using the obtained second deep learning model to recognize the minority-class metallographic structure. According to the method, the recognition accuracy of the model on a small number of categories is remarkably improved by comprehensively using the divide-and-conquer method, the undersampling method and the transfer learning method. The method is easy to implement, and the efficiency and accuracy of the metal material metallographic structure recognition and analysis process can be effectively improved.
Description
technical field [0001] The invention belongs to the technical field of computer vision measurement metallographic detection, and in particular relates to a metallographic structure identification method, system, equipment and storage medium. Background technique [0002] The metallographic phase refers to the physical or chemical state of various components of the metal or alloy inside the alloy, and the metallographic structure reflects the specific shape of the metal phase. Taking the structural material of the pressure device as an example, during the operation of the pressure device, the material is subjected to long-term effects of factors such as temperature, stress, and corrosive media, and its performance will decline, and its organizational structure will change. It needs to be detected and analyzed by metallographic inspection technology. The deterioration of the material is used to evaluate the service safety of the device. One of the main tasks of metallographic...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/00G06K9/62G06N3/04G06N3/08
CPCG06T7/0004G06N3/04G06N3/08G06T2207/20081G06T2207/30136G06F18/241Y02P90/30
Inventor 胡海军申志远李秀峰谢国山韩志远张钰康晓鹏甄宏展李涌泉胡振龙肖尧钱龚雪茹
Owner XI AN JIAOTONG UNIV