A Method for Improving Domain Transformation Phenomena in Image Recognition Models
An image recognition and model technology, applied in the field of computer vision, can solve problems such as insufficient image feature extraction, simple individuals, and affecting the effect of genetic algorithms, and achieve the effect of improving domain transformation, improving complex processes, and increasing complexity
Active Publication Date: 2021-04-27
BEIJING UNIV OF CIVIL ENG & ARCHITECTURE
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] 1. A simple random search method is used in the process of constructing attribute combinations, which cannot ensure a balanced search for all attributes, which affects the distribution of individuals in the genetic algorithm population
[0006] 2. Only 3 attributes are used as an attribute combination, resulting in too simple individuals in the population, which affects the effect of the entire genetic algorithm
[0008] 4. The convolutional neural network model constructed in the whole process is not perfect, which will lead to insufficient feature extraction of the image
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
[0059] Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary only for explaining the present invention and should not be construed as limiting the present invention.
[0060] Those skilled in the art will understand that unless otherwise stated, the singular forms "a", "an", "said" and "the" used herein may also include plural forms. It should be further understood that the word "comprising" used in the description of the present invention refers to the presence of said features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, Integers, steps, operations, elements, components, and / or groups thereof. It will be understoo...
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 provides a method for improving the domain transformation phenomenon of an image recognition model. The genetic algorithm is used to assign a weight value to each attribute, and the weight value of the attribute is continuously updated in the process of searching for the attribute. Calculate the probability of each attribute being searched according to the weight value of the attribute. The smaller the weight value is, the higher the probability of being searched is, otherwise the probability of being searched is smaller. Searching for attributes according to the search probability of each attribute can control the direction of the search, so as to avoid certain attributes being searched too many times or too few times. The combination of the Equal function and the Sum function is used to complete the screening of attribute combinations and improve the complex process of calculation. In the whole process, a convolutional neural network with 3 convolutional layers constructed based on the Leaky Relu activation function is used, which avoids the phenomenon that neurons less than 0 cannot update parameters during the model training process, and achieves the purpose of fully extracting image features. .
Description
technical field [0001] The invention relates to the technical field of computer vision, in particular to a method for improving the domain transformation phenomenon of an image recognition model. Background technique [0002] Image recognition refers to the use of computers to process, analyze and understand images, and is an important field of artificial intelligence. The recognition of the image by the computer is mainly based on the main features of the image, for example, there is a "circle" in the number "6", and there are two "circles" in the number "8". However, images in different fields often have different main features, such as the size of the image, the number of channels, the background color and other features will be generated for the image recognition model. However, in the recognition process, the computer cannot make independent dialectical judgments on the image like the human eye, and cannot directly eliminate the influence of some unnecessary image feat...
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): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/214G06F18/241
Inventor 谭志刘兴业
Owner BEIJING UNIV OF CIVIL ENG & ARCHITECTURE