A bpnn feature recognition method based on improved nba algorithm
A feature recognition and algorithm technology, applied in the field of BPNN feature recognition based on the improved NBA algorithm, can solve the problems that the feature recognition method cannot effectively identify combined features, large feature recognition errors, and large convergence errors
Active Publication Date: 2021-04-09
ZHEJIANG UNIV OF TECH
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] The present invention solves the problem in the prior art that the feature recognition method cannot effectively recognize the combined features, or the error of feature recognition is large, the convergence error is large, or the optimization accuracy is not high, and the problem that the particle is easy to enter the late stage of the iteration is premature. The present invention provides An Optimized BPNN Feature Recognition Method Based on Improved NBA Algorithm
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 Construction
[0064] The present invention will be described in further detail below in conjunction with the examples, but the protection scope of the present invention is not limited thereto.
[0065] The invention relates to a BPNN feature recognition method based on an improved NBA algorithm, and the method includes the following steps.
[0066] Step 1: Preprocess the face-edge adjacency graph, extract the smallest subgraph of eigenfactors, and aggregate the eigenfactors belonging to the same feature into composite features.
[0067] The step 1 includes the following steps.
[0068] Step 1.1: Traverse any face in the face-edge adjacency graph, create a vertex of the attribute adjacency graph corresponding to each face, and extract the attribute of each face as the attribute of the corresponding attribute adjacency graph vertex.
[0069] Step 1.2: For every two faces in the face-edge adjacency graph, identify the adjacency relationship between them, and use the adjacency relationship as ...
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 relates to a BPNN feature recognition method based on the improved NBA algorithm, which preprocesses the face-edge adjacency graph, extracts the minimum subgraph of the feature factor, aggregates the feature factors belonging to the same feature into a composite feature, and performs each feature after the aggregation is completed. The factor is encoded to obtain the feature encoding sequence, and the NBA algorithm is improved by using the second-order oscillation mechanism and the difference algorithm, and the improved NBA algorithm is used to optimize the BP neural network and perform feature recognition. The present invention identifies the features with engineering significance to the greatest extent, and because the neural network has excellent learning performance, the accuracy and efficiency of feature recognition are greatly improved, and the improved NBA algorithm is used to optimize the BP neural network, which can realize the control of local The mutual conversion between search and global search avoids falling into local optimal defects and has better convergence. The invention performs feature recognition after training, which effectively improves the accuracy and efficiency of feature recognition.
Description
technical field [0001] The present invention relates to the technical field for reading or recognizing printed or written characters or for recognizing graphics, in particular to a BPNN feature recognition method based on an improved NBA algorithm that uses composite processing feature extraction to maximize the identification of features with engineering significance . Background technique [0002] Feature recognition technology is constantly updated with the development of science and technology. Experts and scholars in the scientific community have made great progress in feature recognition technology through continuous research and algorithm update through years of hard work, and various methods emerge in endlessly. [0003] However, the existing feature recognition technology still has many problems that are difficult to solve or the solution effect is not satisfactory. Among them, the graph-based method is the most widely used method, but this method has a large amoun...
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 Patents(China)
IPC IPC(8): G06N3/08G06K9/62
CPCG06N3/084G06N3/086G06F18/214
Inventor 简琤峰林崇李苗张美玉
Owner ZHEJIANG UNIV OF TECH