Image classification method, apparatus and computer equipment
A classification method and image technology, applied in the field of image processing, can solve problems such as the decline in the accuracy of image classification, and achieve the effect of improving the accuracy and avoiding the impact.
Active Publication Date: 2022-07-01
CHANGSHA UNIVERSITY OF SCIENCE AND TECHNOLOGY
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] Based on this, it is necessary to address the above-mentioned technical problems and provide an image classification method, device and computer equipment that can solve the problem that the accuracy of image classification will decrease due to too small background information during average pooling
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
[0036] In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application will be described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application.
[0037] In one embodiment, as figure 1 As shown, an image classification method is provided, including the following steps:
[0038] Step 102: Obtain an image sample, perform k minimum pooling on the data in each pooling frame in the image sample through the first pooling layer, obtain k minimum pooling feature values, and take the average value to obtain the minimum pooling average value .
[0039] Among them, k is the dimension of the pooling frame. When performing pooling processing, the data in the pooling frame is processed. For example, when performing maximum p...
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 present application relates to an image classification method, apparatus and computer equipment. The method includes: acquiring image samples, performing k minimum pooling on each pooled frame in the convolved feature map through a first pooling layer, and averaging the k minimum values to obtain a minimum pooling average. value; average pooling of the image samples through the second pooling layer to obtain the average pooling feature map; according to the preset weight, weight the average pooling feature map to obtain the extreme threshold value corresponding to each pooling frame ; If the average value of the minimum pooling is less than the extreme threshold, discard the minimum value in the pooling frame and perform average pooling to obtain the de-extreme feature map corresponding to the image sample; use the de-extreme feature map to train the image classification model to get the training A good image classification model; obtain the de-extended feature map of the image to be classified, and input the de-extended feature map into the trained image classification model to obtain the image category. The method can improve the accuracy of image classification.
Description
technical field [0001] The present application relates to the technical field of image processing, and in particular, to an image classification method, apparatus and computer equipment. Background technique [0002] When classifying an image, it is usually classified by a convolutional neural network. In a convolutional neural network, the image is first pooled. The most common pooling methods are max pooling and average pooling. Max pooling only retains the maximum value in the pooling box, so max pooling can effectively extract the most representative information in the feature map. Average pooling calculates the mean of all values in the pooling box, so that all the information in the feature map can be obtained on average without losing too much key information. These two methods have been widely used in previous deep neural network models because of their simple calculation, good effect, and can solve the problems of reducing parameters and dimensions. However, the ...
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): G06V10/764G06V10/774G06V10/82G06K9/62G06N3/04
CPCG06N3/045G06F18/241G06F18/214
Inventor 邓泽林秦平越
Owner CHANGSHA UNIVERSITY OF SCIENCE AND TECHNOLOGY