CN113077525A - Image classification method based on frequency domain contrast learning - Google Patents

Image classification method based on frequency domain contrast learning Download PDF

Info

Publication number
CN113077525A
CN113077525A CN202110164693.8A CN202110164693A CN113077525A CN 113077525 A CN113077525 A CN 113077525A CN 202110164693 A CN202110164693 A CN 202110164693A CN 113077525 A CN113077525 A CN 113077525A
Authority
CN
China
Prior art keywords
image
frequency domain
learning
training
features
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110164693.8A
Other languages
Chinese (zh)
Inventor
袁召全
邵焕
吴晓
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southwest Jiaotong University
Original Assignee
Southwest Jiaotong University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Southwest Jiaotong University filed Critical Southwest Jiaotong University
Priority to CN202110164693.8A priority Critical patent/CN113077525A/en
Publication of CN113077525A publication Critical patent/CN113077525A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/90Determination of colour characteristics
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10Complex mathematical operations
    • G06F17/14Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
    • G06F17/141Discrete Fourier transforms
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10Complex mathematical operations
    • G06F17/16Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/21Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology
    • G06N3/045Combinations of networks
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/10Segmentation; Edge detection
    • G06T7/11Region-based segmentation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Databases & Information Systems (AREA)
  • Evolutionary Computation (AREA)
  • Algebra (AREA)
  • Computing Systems (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Health & Medical Sciences (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Discrete Mathematics (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

一种基于频域对比学习的图像分类方法,包括以下步骤:S1:对训练集图像,分别做两次随机的数据增强,对同一张图像做两次随机数据增强会而到两张不同的增强后的图像;S2:将数据增强后的图像做离散余弦变换,得到转移到频域的图像;S3:将转移到频域的图像通过深度网络,通过对比学习任务,学习网络参数,得到图像的稳定特征;S5:利用参数已经优化的深度网络,对测试集中的图像进行分类。本发明能够学习到同一类物体在不同背景下的稳定特征,比传统的分类方法能更好地进行非独立同分布图像的分类。

Figure 202110164693

An image classification method based on frequency domain contrastive learning, comprising the following steps: S1: Perform two random data enhancements on the training set images respectively, and perform two random data enhancements on the same image, resulting in two different enhancements S2: Discrete cosine transform the image after data enhancement to obtain the image transferred to the frequency domain; S3: Pass the image transferred to the frequency domain through a deep network, and learn network parameters by comparing the learning tasks to obtain the image Stabilizing features; S5: Classify images in the test set using a deep network whose parameters have been optimized. The invention can learn the stable features of the same type of objects under different backgrounds, and can better classify non-IID images than traditional classification methods.

Figure 202110164693

Description

Image classification method based on frequency domain contrast learning
Technical Field
The invention relates to the field of image classification, in particular to an image classification method based on frequency domain contrast learning.
Background
In recent years, with the continuous development of deep learning techniques, computers have achieved considerably high precision in various image classification tasks. However, the effectiveness of the mainstream image classification method is established on the premise that the training data set and the test set satisfy independent and same distribution. In practical application, training data and test data are often difficult to strictly satisfy the assumption of independent and same distribution, and in such a case, a model trained by a traditional method can perform well on a training data set, but an ideal classification effect cannot be achieved on the test data.
The classification methods proposed for the non-independent images with the same distribution are not many, and the currently proposed method is mainly based on a causal inference theory. The method inputs images into a depth model to extract features, takes each dimension of the features as an intervention variable in turn, and takes the features of other dimensions as confounding factors. The model is made to learn a set of sample weights, minimizing the association between each dimensional feature, thereby independently estimating the causal relationship of each dimensional feature to the classification result. However, there is not at all a complete lack of association between each dimension of an image feature, and different dimensions of features have different causal relationships to image classes. The method treats each dimension feature equally, and the classification effect is still not ideal enough.
The non-independent and identical distribution of the training data and the test data is mainly caused by different context information (including the background of the classification target, the texture of the classification target, the action of the classification target and the like) in the image, but the classification target has some characteristics which are not changed under different scenes. The stable characteristics of the learning target can effectively solve the classification problem of the non-independent same-distribution image. However, the diversity of the target stabilization features and the abstraction of the high-dimensional features extracted by the neural network pose challenges to the learning of the stabilization features.
Disclosure of Invention
Objects of the invention
In order to solve the technical problems in the background art, the invention provides an image classification method based on frequency domain contrast learning.
(II) technical scheme
In order to solve the above problems, the present invention provides an image classification method based on frequency domain contrast learning, which comprises the following steps:
s1: respectively carrying out random data enhancement twice on the training set image, wherein the data enhancement operation comprises cutting and size adjustment, horizontal turning, Gaussian blur, color dithering and gray level image conversion;
whether each data enhancement operation is executed is determined by the probability set in advance so as to carry out two times of random data enhancement on the same image to obtain two different enhanced images;
s2: performing discrete cosine transform on the image subjected to data enhancement to obtain an image transferred to a frequency domain;
wherein, the image is represented by RGB color table codes; extracting the frequency domain features of the image can be divided into the following two sub-steps:
s201: converting an image from an RGB color space to a YCbCr color space according to the following formula:
Figure BDA0002937434440000021
202: the picture converted to YCbCr color space is divided into 8 x 8 small blocks, and three channels are obtained for each block according to the formula F-AfATSolving a corresponding discrete cosine transform coefficient; the transformation matrix formula is as follows:
Figure BDA0002937434440000031
the original image is divided into 14 small blocks of 8 × 8, each small block contains 64 pixel points, and each pixel point has pixel values of 3 color channels, that is, 192 frequency domain coefficients can be obtained for each image small block, so that image frequency domain coefficients with the dimensions of (192, 14, 14) are obtained;
s3: the image transferred to the frequency domain passes through a depth network, and network parameters are learned through a comparison learning task to obtain the stable characteristics of the image; the method comprises the following substeps:
s301: respectively inputting the image frequency domain coefficients into a feature extraction layer of the depth network to obtain a feature h with the dimension of (N, 2048)iAnd hjThe network structure adopts a residual error neural network;
s302: h is to beiAnd hjInputting into a multi-layer perceptron to obtain an (N, 128) -dimensional feature z for comparison learning trainingiAnd zj
S303: will ziAnd zjAnd (3) splicing according to the 0 th dimension to obtain the characteristics for calculating the comparative learning loss:
Figure BDA0002937434440000032
the loss of the comparative learning pre-training is calculated according to the following formula:
Figure BDA0002937434440000033
Figure BDA0002937434440000034
Figure BDA0002937434440000035
wherein σ is a positive number;
s304: performing parameter adjustment on the depth network in return by minimizing the loss of the contrast learning pre-training, and performing global parameter adjustment by adopting a back propagation algorithm until the loss of the contrast learning pre-training does not decrease any more, then converging the model, and ending the contrast learning pre-training step;
s4: predicting the classification result of the training set image by using the extracted features, further learning network parameters, and performing classification tasks;
wherein the 2018-dimensional features extracted in S301 are input into the full link layer and the softmax layerIn the classifier, a predicted classification result is obtained:
Figure BDA0002937434440000041
n is the number of images in a batch, K is the number of categories of images;
and then calculating a cross entropy loss function by using the classification result:
Figure BDA0002937434440000042
and finally, carrying out global parameter adjustment through a back propagation algorithm, and optimizing the network parameters by taking the minimized cross entropy loss function as a target until the function value is not reduced any more.
S5: and classifying the images in the test set by using the depth network with optimized parameters:
performing random data enhancement on the test image according to S1, and converting the test image into a frequency domain according to S2; and finally, inputting the image converted into the frequency domain into a feature extraction layer of the depth network, and inputting the image into a classifier consisting of a full connection layer and a softmax layer to obtain a prediction result.
Preferably, in S2, the frequency domain image is input into a convolutional neural network to extract features.
Preferably, in S3, the model learns stable features through pre-training of the contrast learning task in combination with the frequency domain learning and contrast learning framework.
In the invention, images of each category are classified secondarily according to context information, and then relevant data sets are divided: training set: the image classification method comprises an image and a classification label corresponding to the image. And (3) test set: the image classification method comprises an image and a classification label corresponding to the image. But the context information of the images in the test set is different from that in the training set.
In the invention, random data enhancement is performed twice on each batch of pictures, and then the pictures are converted into the frequency domain to obtain two characteristics of the same image, and the training model is used for distinguishing whether the two characteristics come from the same image, so that the stable characteristics of the image are learned, and the classification effect of the non-independent images in the same distribution is improved.
The method can learn the stable characteristics of the same object under different backgrounds, and can better classify the non-independent images with the same distribution compared with the traditional classification method.
Drawings
FIG. 1 is a flowchart of a frequency domain contrast learning-based non-independent image classification method according to the present invention.
Fig. 2 is a model structure diagram of frequency domain contrast learning in the image classification method based on frequency domain contrast learning according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the accompanying drawings in conjunction with the following detailed description. It should be understood that the description is intended to be exemplary only, and is not intended to limit the scope of the present invention. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present invention.
As shown in fig. 1-2, the image classification method based on frequency domain contrast learning provided by the present invention includes the following steps:
s1: for training set image [ x1,x2,x3,……xN]Respectively carrying out random data enhancement twice, wherein the data enhancement operation comprises cutting and size adjustment, horizontal turning, Gaussian blur, color dithering and gray level image conversion;
whether each data enhancement operation is executed is determined by the probability set in advance so as to carry out two times of random data enhancement on the same image to obtain two different enhanced images;
s2: performing discrete cosine transform on the image subjected to data enhancement to obtain an image transferred to a frequency domain, namely:
[xi1,xi2,xi3,……xiN]and [ x ]j1,xj2,xj3,……xjN];
Wherein, the image is represented by RGB color table codes;
the enhanced image dimension is (3, 112, 112), where 3 represents that the image has 3 color channels, which are R, G respectively, and B, 112 is the image size set in advance;
extracting the frequency domain features of the image can be divided into the following two sub-steps:
s201: converting an image from an RGB color space to a YCbCr color space according to the following formula:
Figure BDA0002937434440000061
202: the picture converted to YCbCr color space is divided into 8 x 8 small blocks, and three channels are obtained for each block according to the formula F-AfATSolving a corresponding discrete cosine transform coefficient; the transformation matrix formula is as follows:
Figure BDA0002937434440000062
the original picture (3, 112, 112) can be divided into 14 8 × 8 small blocks, each small block contains 64 pixel points, and each pixel point has a pixel value of 3 color channels, that is, 192 frequency domain coefficients can be obtained for each image small block, so as to obtain an image frequency domain coefficient with a dimension of (192, 14, 14):
[x’i1,x’i2,x’i3,……x’iN]and [ x'j1,x’j2,x’j3,……x’jN];
S3: the image transferred to the frequency domain passes through a depth network, and network parameters are learned through a comparison learning task to obtain the stable characteristics of the image; the method comprises the following substeps:
s301: image frequency domain coefficient [ x'i1,x’i2,x’i3,……x’iN]And [ x'j1,x’j2,x’j3,……x’jN]Respectively inputting the feature extraction layers of the depth network to obtain a feature h with dimensions of (N, 2048)iAnd hjThe network structure thereof adoptsUsing a residual neural network;
s302: h is to beiAnd hjInputting into a multi-layer perceptron to obtain an (N, 128) -dimensional feature z for comparison learning trainingiAnd zj
S303: will ziAnd zjAnd (3) splicing according to the 0 th dimension to obtain the characteristics for calculating the comparative learning loss:
Figure BDA0002937434440000071
the loss of the comparative learning pre-training is calculated according to the following formula:
Figure BDA0002937434440000072
Figure BDA0002937434440000073
Figure BDA0002937434440000074
wherein σ is a positive number;
s304: performing parameter adjustment on the depth network in return by minimizing the loss of the contrast learning pre-training, and performing global parameter adjustment by adopting a back propagation algorithm until the loss of the contrast learning pre-training does not decrease any more, then converging the model, and ending the contrast learning pre-training step;
s4: predicting the classification result of the training set image by using the extracted features, further learning network parameters, and performing classification tasks;
inputting the 2018 dimensional features extracted in the S301 into a classifier consisting of a full-link layer and a softmax layer to obtain a predicted classification result:
Figure BDA0002937434440000075
n is the number of images in a batchMesh, K is the number of categories of the image;
and then calculating a cross entropy loss function by using the classification result:
Figure BDA0002937434440000076
and finally, carrying out global parameter adjustment through a back propagation algorithm, and optimizing the network parameters by taking the minimized cross entropy loss function as a target until the function value is not reduced any more.
S5: and classifying the images in the test set by using the depth network with optimized parameters:
performing random data enhancement on the test image according to S1, and converting the test image into a frequency domain according to S2; and finally, inputting the image converted into the frequency domain into a feature extraction layer of the depth network, and inputting the image into a classifier consisting of a full connection layer and a softmax layer to obtain a prediction result.
In the invention, images of each category are classified secondarily according to context information, and then relevant data sets are divided: training set: the image classification method comprises an image and a classification label corresponding to the image. And (3) test set: the image classification method comprises an image and a classification label corresponding to the image. But the context information of the images in the test set is different from that in the training set.
In the invention, random data enhancement is performed twice on each batch of pictures, and then the pictures are converted into the frequency domain to obtain two characteristics of the same image, and the training model is used for distinguishing whether the two characteristics come from the same image, so that the stable characteristics of the image are learned, and the classification effect of the non-independent images in the same distribution is improved.
In an alternative embodiment, in S2, the frequency domain image is input into a convolutional neural network to extract features.
In an alternative embodiment, in S3, the model is made to learn stable features through pre-training of the contrast learning task in conjunction with the frequency domain learning and contrast learning framework.
In conclusion, the method and the device can learn the stable characteristics of the same object under different backgrounds, and can better classify the non-independent images with the same distribution compared with the traditional classification method.
It is to be understood that the above-described embodiments of the present invention are merely illustrative of or explaining the principles of the invention and are not to be construed as limiting the invention. Therefore, any modification, equivalent replacement, improvement and the like made without departing from the spirit and scope of the present invention should be included in the protection scope of the present invention. Further, it is intended that the appended claims cover all such variations and modifications as fall within the scope and boundaries of the appended claims or the equivalents of such scope and boundaries.

Claims (3)

1.一种基于频域对比学习的图像分类方法,其特征在于,包括以下步骤:1. an image classification method based on frequency domain contrastive learning, is characterized in that, comprises the following steps: S1:对训练集图像,分别做两次随机的数据增强,数据增强操作包括裁剪调整大小、水平翻转、高斯模糊、色彩抖动、转灰度图;S1: Perform two random data enhancements on the images of the training set. The data enhancement operations include cropping and resizing, horizontal flipping, Gaussian blurring, color dithering, and grayscale conversion; 每项数据增强操作是否执行由事先设置的概率决定的,以对同一张图像做两次随机数据增强会而到两张不同的增强后的图像;Whether each data enhancement operation is performed is determined by a preset probability, so that two random data enhancements on the same image will result in two different enhanced images; S2:将数据增强后的图像做离散余弦变换,得到转移到频域的图像;S2: Perform discrete cosine transform on the image after data enhancement to obtain an image transferred to the frequency domain; 其中,图像采用RGB颜色表代码表示;提取图像的频域特征可以分为以下两个子步骤:Among them, the image is represented by RGB color table code; extracting the frequency domain features of the image can be divided into the following two sub-steps: S201:依照下列公式将图像从RGB颜色空间转换到YCbCr颜色空间:S201: Convert the image from the RGB color space to the YCbCr color space according to the following formula:
Figure FDA0002937434430000011
Figure FDA0002937434430000011
202:将转换到YCbCr颜色空间的图片分割成8×8的若干个小块,对每一块求其三个频道,根据公式F=AfAT求对应的离散余弦变换系数;转换矩阵公式如下:202: Divide the picture converted into the YCbCr color space into several small blocks of 8×8, find three channels for each block, and find the corresponding discrete cosine transform coefficient according to the formula F= AfAT ; the conversion matrix formula is as follows:
Figure FDA0002937434430000012
Figure FDA0002937434430000012
其中,原图片划分为14个8×8的小块,每个小块含有有64个像素点,每个像素点有3个颜色通道的像素值,即,每个图像小块可以求得192个频域系数,从而得到维度为(192,14,14)的图像频域系数;Among them, the original image is divided into 14 8×8 small blocks, each small block contains 64 pixels, and each pixel has the pixel value of 3 color channels, that is, each image small block can be calculated as 192 frequency domain coefficients, thereby obtaining image frequency domain coefficients with dimensions (192, 14, 14); S3:将转移到频域的图像通过深度网络,通过对比学习任务,学习网络参数,得到图像的稳定特征;其包括以下几个子步骤:S3: Pass the image transferred to the frequency domain through a deep network, and learn the network parameters by comparing the learning tasks to obtain the stable features of the image; it includes the following sub-steps: S301:将图像频域系数分别输入深度网络的特征提取层,得到维度为(N,2048)的特征hi和hj,其网络结构采用残差神经网络;S301: Input the image frequency domain coefficients into the feature extraction layer of the deep network, respectively, to obtain features h i and h j with dimensions (N, 2048), and the network structure adopts a residual neural network; S302:将hi和hj输入多层感知机,得到用于对比学习训练的(N,128)维特征zi和zjS302: Input h i and h j into a multilayer perceptron to obtain (N, 128)-dimensional features z i and z j for comparative learning and training; S303:将zi和zj按第0维进行拼接,得到用于计算对比学习损失的特征:S303: splicing z i and z j according to the 0th dimension to obtain the features used to calculate the contrastive learning loss:
Figure FDA0002937434430000021
Figure FDA0002937434430000021
依据下列公式计算对比学习预训练的损失:The loss of contrastive learning pre-training is calculated according to the following formula:
Figure FDA0002937434430000022
Figure FDA0002937434430000022
Figure FDA0002937434430000023
Figure FDA0002937434430000023
Figure FDA0002937434430000024
Figure FDA0002937434430000024
其中,σ为正数;Among them, σ is a positive number; S304:通过最小化上述对比学习预训练的损失,反过来对深度网络进行参数调整,采用后向传播算法进行全局参数调整,直到对比学习预训练的损失不再下降,则模型收敛,对比学习预训练步骤结束;S304: By minimizing the loss of the pre-training of the comparative learning, in turn, adjust the parameters of the deep network, and use the back-propagation algorithm to adjust the global parameters. Until the loss of the pre-training of the comparative learning does not decrease, the model converges, and the comparative learning is pre-trained. The training step is over; S4:对训练集图像,利用提取的特征预测其分类结果,进一步学习网络参数,进行分类任务;S4: For the training set images, use the extracted features to predict the classification results, further learn network parameters, and perform classification tasks; 其中,将S301中提取的2018维特征输入全连接层和softmax层组成的分类器中,得到预测的分类结果:
Figure FDA0002937434430000025
N为一个批次中图像的数目,K为图像的类别数目,;
Among them, the 2018-dimensional features extracted in S301 are input into the classifier composed of the fully connected layer and the softmax layer, and the predicted classification result is obtained:
Figure FDA0002937434430000025
N is the number of images in a batch, K is the number of categories of images,;
再利用分类结果计算交叉熵损失函数:Then use the classification result to calculate the cross entropy loss function:
Figure FDA0002937434430000031
Figure FDA0002937434430000031
最后,通过后向传播算法进行全局参数调整,以最小化交叉熵损失函数为目标优化网络参数,直到该函数值不再下降;Finally, the global parameter adjustment is carried out through the back propagation algorithm, and the network parameters are optimized to minimize the cross entropy loss function until the value of the function no longer decreases; S5:利用参数已经优化的深度网络,对测试集中的图像进行分类:S5: Classify images in the test set using a deep network with optimized parameters: 对测试图像做按照S1中的随机数据增强,再按照S2转换到频域;最后,将转换到频域后的图像输入深度网络的特征提取层,并输入由全连接层和softmax层组成的分类器中,得到预测结果。The test image is enhanced according to the random data in S1, and then converted to the frequency domain according to S2; finally, the image converted to the frequency domain is input into the feature extraction layer of the deep network, and the classification consisting of the fully connected layer and the softmax layer is input. , and get the prediction result.
2.根据权利要求1所述的基于频域对比学习的图像分类方法,其特征在于,在S2中,将频域图像输入卷积神经网络中提取特征。2 . The image classification method based on frequency domain contrastive learning according to claim 1 , wherein, in S2 , the frequency domain image is input into a convolutional neural network to extract features. 3 . 3.根据权利要求1所述的基于频域对比学习的图像分类方法,其特征在于,在S3中,结合频域学习和对比学习框架,通过对比学习任务的预训练使模型学习稳定的特征。3. The image classification method based on frequency domain contrastive learning according to claim 1, is characterized in that, in S3, combined with frequency domain learning and contrastive learning framework, through the pre-training of contrastive learning task, make the model learn stable features.
CN202110164693.8A 2021-02-06 2021-02-06 Image classification method based on frequency domain contrast learning Pending CN113077525A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110164693.8A CN113077525A (en) 2021-02-06 2021-02-06 Image classification method based on frequency domain contrast learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110164693.8A CN113077525A (en) 2021-02-06 2021-02-06 Image classification method based on frequency domain contrast learning

Publications (1)

Publication Number Publication Date
CN113077525A true CN113077525A (en) 2021-07-06

Family

ID=76609296

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110164693.8A Pending CN113077525A (en) 2021-02-06 2021-02-06 Image classification method based on frequency domain contrast learning

Country Status (1)

Country Link
CN (1) CN113077525A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114357221A (en) * 2022-03-15 2022-04-15 南京航空航天大学 Self-supervision active learning method based on image classification

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020186765A1 (en) * 2001-06-05 2002-12-12 Morley Steven A. Selective chrominance decimation for digital images
CN108009493A (en) * 2017-11-30 2018-05-08 电子科技大学 Face anti-fraud recognition methods based on action enhancing
CN111127360A (en) * 2019-12-20 2020-05-08 东南大学 Gray level image transfer learning method based on automatic encoder
CN111784633A (en) * 2020-05-26 2020-10-16 西安理工大学 An automatic detection algorithm for insulator defects for power inspection video

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020186765A1 (en) * 2001-06-05 2002-12-12 Morley Steven A. Selective chrominance decimation for digital images
CN108009493A (en) * 2017-11-30 2018-05-08 电子科技大学 Face anti-fraud recognition methods based on action enhancing
CN111127360A (en) * 2019-12-20 2020-05-08 东南大学 Gray level image transfer learning method based on automatic encoder
CN111784633A (en) * 2020-05-26 2020-10-16 西安理工大学 An automatic detection algorithm for insulator defects for power inspection video

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
HUAN SHAO,AND ETC: "Contrastive Learning in Frequency Domain for Non-I.I.D.Image Classification", 《SPRINGER LINK:INTERNATIONAL CONFERENCE ON MULTIMEDIA MODELING》 *
张晶主编: "《多媒体信息与通信》", 30 November 2017, 西安:西安电子科技大学出版社 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114357221A (en) * 2022-03-15 2022-04-15 南京航空航天大学 Self-supervision active learning method based on image classification
CN114357221B (en) * 2022-03-15 2022-08-05 南京航空航天大学 Self-supervision active learning method based on image classification

Similar Documents

Publication Publication Date Title
CN109711413B (en) Image semantic segmentation method based on deep learning
CN109299274B (en) Natural scene text detection method based on full convolution neural network
CN111639692B (en) Shadow detection method based on attention mechanism
AU2019451948B2 (en) Real-time video ultra resolution
CN111046962A (en) Sparse attention-based feature visualization method and system for convolutional neural network model
JP7381942B2 (en) Control method, information processing device and control program
CN111047543B (en) Image enhancement method, device and storage medium
CN111986142A (en) A method for unsupervised enhancement of image data of surface defects of hot-rolled coils
CN109348211B (en) A general information hiding detection method for video intra- and inter-coding
CN109344893B (en) A kind of image classification method based on mobile terminal
CN107784288A (en) A kind of iteration positioning formula method for detecting human face based on deep neural network
CN110929099B (en) Short video frame semantic extraction method and system based on multi-task learning
CN113837965B (en) Image definition identification method and device, electronic equipment and storage medium
CN108710893A (en) A kind of digital image cameras source model sorting technique of feature based fusion
CN114897884A (en) No-reference screen content image quality evaluation method based on multi-scale edge feature fusion
CN115240024A (en) Method and system for segmenting extraterrestrial pictures by combining self-supervised learning and semi-supervised learning
CN118229569A (en) Document image transmission removal method and device based on fuzzy diffusion model
CN116563410A (en) Electrical equipment electric spark image generation method based on two-stage generation countermeasure network
CN117689996A (en) Model generation method and device, electronic equipment and storage medium
CN116935125A (en) Noise data set target detection method realized through weak supervision
Lee et al. SAF-Nets: Shape-adaptive filter networks for 3D point cloud processing
CN116681627A (en) A Cross-Scale Fusion Adversarial Enhancement Method for Adaptive Underwater Image Generation
CN113077525A (en) Image classification method based on frequency domain contrast learning
CN118864816B (en) Multi-source passive field adaptive target detection method and system based on deep learning
Ma et al. A novel generative image inpainting model with dense gated convolutional network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20210706

RJ01 Rejection of invention patent application after publication