CN114444565A - Image tampering detection method, terminal device and storage medium - Google Patents
Image tampering detection method, terminal device and storage mediumDownload PDFInfo
- Publication number
- CN114444565A CN114444565ACN202111533371.2ACN202111533371ACN114444565ACN 114444565 ACN114444565 ACN 114444565ACN 202111533371 ACN202111533371 ACN 202111533371ACN 114444565 ACN114444565 ACN 114444565A
- Authority
- CN
- China
- Prior art keywords
- image
- layer
- loss function
- tampering detection
- input
- 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.)
- Granted
Links
Images
Classifications
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/29—Graphical models, e.g. Bayesian networks
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Storage Device Security (AREA)
- Image Analysis (AREA)
Abstract
Description
Technical Field
The present invention relates to the field of image detection, and in particular, to an image tampering detection method, a terminal device, and a storage medium.
Background
With the development of science and technology and the popularization of software, the operation of producing high-quality and vivid forged pictures by PS becomes simpler and more convenient. Many high-quality P pictures can cheat human eyes and detection models successfully, which causes great harm to social security development.
The basic operations of image PS tampering can be divided into four categories, namely copy-move (copy-move), split-stitch (stitching), remove and fill-in with other pixels (remove), and generate characters (generated). In order to increase the fidelity of the forged image, a falsifier not only uses tools to perform operations such as edge feathering, painting brush painting, color change and the like, but also performs post-processing operations such as noise adding, blurring adding, compression and the like. These operations make it difficult to distinguish the image features of the forged region from the non-tampered region, and increase the difficulty of detection to some extent.
The method for identifying and positioning the PS tampering is mainly divided into a traditional identification method and a deep learning-based method. The traditional detection and positioning algorithm is an algorithm for detecting and positioning based on the statistical characteristics of images according to the self artificial structural characteristics, for example, the detection and positioning algorithm based on DCT counterfeiting tampering detection algorithm, SIFT detection algorithm, resampling trace detection algorithm and the like, and the detection and positioning algorithm based on deep learning autonomously learns the relevant characteristics so as to detect the PS tampered image area.
The detection and positioning method based on deep learning has been developed for a while, but with the development of image tampering technology, the current tampering detection algorithm has some defects, for example, most of the tampering detection and positioning algorithms only extract features based on RGB format images, and the extracted features are limited, so that images in a forged region and an untampered region are difficult to distinguish.
Disclosure of Invention
In order to solve the above problems, the present invention provides an image tampering detection method, a terminal device, and a storage medium.
The specific scheme is as follows:
an image tampering detection method, comprising the steps of:
s1: acquiring images of two types, namely tampered images and images of non-tampered images to form a training set, and labeling tampered areas in the tampered images in the training set;
s2: constructing an image tampering detection model, and training the model through a training set;
a trunk part in a network structure of the image tampering detection model adopts a fast-RCNN network, after an input image is input into the fast-RCNN network, a candidate frame output by an RPN layer in the fast-RCNN network is input into an ROI posing layer in the fast-RCNN network on one hand, and a new ROI posing layer is input together with the input image on the other hand, an image output by the ROI posing layer is resampled to obtain a resampled spectrogram, the resampled spectrogram is input into an LSTM network to obtain a resampled characteristic diagram, the output of the resampled characteristic diagram after being subjected to an upsampling layer and a convolutional layer is input into a bilinear pooling layer together with the output of the fast-RCNN network to be fused, and the fused result is subjected to a full connection layer and a softmax layer to obtain a prediction result of whether the image is tampered or not;
s3: and detecting whether the image is tampered through the trained model.
Further, the resampling process comprises: and processing the candidate frame region in the image through an EM algorithm, and then performing Fourier transform to obtain a resample spectrogram.
Further, a loss function L of the image tampering detection modeltotalThe calculation formula of (2) is as follows:
Ltotal＝LRPN+Ltamper+Lbbox
wherein L isRPNRepresenting the loss function, L, of the RPN layertamperRepresents the final classification loss function, LbboxA rectangular box regression loss function is represented.
Further, the loss function of the RPN layer includes a classification loss function LclsAnd a regression loss function LregThe formula for calculating the loss function of the RPN layer is:
wherein, giRepresenting the predicted classification probability of the candidate box,
true tag representing candidate box, fiThe coordinates representing the prediction of the candidate box,
representing the real coordinates of the labeled frame corresponding to the candidate frame, i representing the serial number of the candidate frame, λ representing the weight coefficient, NclsIndicating the size of the small batch, NregIndicating the number of candidate boxes.
Further, the final classification loss function adopts a cross entropy loss function, and the rectangular box regression loss function adopts a SmoothL1 loss function.
An image tampering detection terminal device comprises a processor, a memory and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to realize the steps of the method of the embodiment of the invention.
A computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, carries out the steps of the method as described above for an embodiment of the invention.
According to the technical scheme, the resampling characteristics are used in the LSTM, the relation between the tampered blocks and the non-tampered blocks is explored, and the problem that the characteristic extraction of the current deep learning network in the field of counterfeiting and tampering is insufficient is solved.
Drawings
Fig. 1 is a flowchart illustrating a first embodiment of the present invention.
Fig. 2 is a schematic network structure diagram of the image tampering detection model in this embodiment.
Detailed Description
To further illustrate the various embodiments, the invention provides the accompanying drawings. The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the embodiments. Those skilled in the art will appreciate still other possible embodiments and advantages of the present invention with reference to these figures.
The invention will now be further described with reference to the accompanying drawings and detailed description.
The first embodiment is as follows:
an embodiment of the present invention provides an image tampering detection method, as shown in fig. 1, the method includes the following steps:
s1: and acquiring images of two types, namely tampered images and images of non-tampered images to form a training set, and labeling tampered areas in the tampered images in the training set.
S2: and constructing an image tampering detection model, and training the model through a training set.
As shown in fig. 2, the main part of the network structure of the image tampering detection model adopts the fast-RCNN network, and after the input image is input into the fast-RCNN network, the candidate frames output by the RPN layer in the fast-RCNN network are input into the ROI posing layer in the fast-RCNN network on one hand to complete the part of the original fast-RCNN network, that is, the coordinates of the candidate frame are mapped to the feature map through the ROI posing layer, on the other hand, a new ROI posing layer is input together with the input image, the coordinates of the candidate frame are corresponding to an input image, the image output by the ROI posing layer is subjected to resampling processing to obtain a resampled spectrogram, the resampled spectrogram is input into an LSTM network to obtain a resampled feature map, the output of the resampled feature map after passing through an upper sampling layer and a convolution layer and the output of a Faster-RCNN network are jointly input into a bilinear pooling layer to be fused, and the fused result passes through a full connection layer and a softmax layer to obtain a prediction result of whether the image is tampered.
The input image in this embodiment is an RGB image.
The resampling process in this embodiment is as follows: and processing the candidate frame region in the image through an EM algorithm, and then performing Fourier transform to obtain a resample spectrogram. The specific process is as follows:
1. first assume that each sample point isBelonging to two models, the first model is an M1 model, and the M1 model represents the currently investigated point yiThere is a linear correlation with its neighbors, that is, the generation of these points is according to the following model:
where n (i) is a gaussian distribution of the independent homographic zero mean unknown variance σ. The second model is the M2 model, the M2 model representing that the sample points under investigation do not have a linear correlation with their neighbors, i.e. they are generated by an independent process.
2. The EM algorithm is a two-step iterative algorithm: (1) e, step E: the probability of the model to which each sample point belongs is estimated first; (2) and M: a special form of correlation of the sample points is estimated. Sampling point yiThe probability of the M1 model can be obtained by a Bayesian formula:
in the above formula Pr { yi∈M1And Pr { y }i∈M2Is a priori probability, with an initial value of 1/2, and y, in additioniThe probability of belonging to the M1 model can be expressed as:
in step M, the estimate for the weight vector is calculated by the following formula:
(3) the numerator term in the equation can be expressed as the prediction residual error, i.e.:
after the calculation of the step E is finished, the weight vector in the step M is calculated
The calculation formula of (2) is as follows:
wherein the row vector of the Y matrix is 2 adjacent to each point in the original image matrixNA vector of points, W is a diagonal matrix, and the values on the diagonal of the matrix are the values of ω (i), the weight vector calculated by this equation
And E and M steps are iterated repeatedly until convergence, and a final weight vector can be obtained at the moment.
3. And then carrying out Fourier transform on the posterior probability map to obtain a resampled spectrogram, wherein the purpose of Fourier transform is to take out high-frequency components with periodicity and remove low-frequency components through a high-pass filter. After the Fourier transformation, if the image contains relevant information of periodicity, the Fourier transformation image contains bright spots of periodicity.
Loss function L of image tampering detection modeltotalThe calculation formula of (2) is as follows:
Ltotal＝LRPN+Ltamper+Lbbox
wherein L isRPNRepresenting the loss function, L, of the RPN layertamperRepresents the final classification loss function, LbboxA rectangular box regression loss function is represented.
The RPN layer has two tasks, which are to judge which candidate frames contain the picture main body and which only contain the background from a plurality of candidate frames, namely a classification task; for candidate frames containing a subject, regression obtains the true goal, i.e., regression task, and thus, in this embodiment, sets the loss packet of the RPN layerIncluding classification loss function LclsAnd a regression loss function LregThe formula for calculating the loss function of the RPN layer is:
wherein, giRepresenting the predicted classification probability of the candidate box (anchor),
true label, f, representing a candidate box (anchor)iRepresenting the coordinates of the candidate box (anchor) prediction,
represents the real coordinate of the label box (ground channel) corresponding to the candidate box (anchor), i represents the number of the candidate box, lambda represents the weight coefficient, NclsIndicates the size of the minibatch (mini-batch size), NregRepresenting the number of candidate boxes.
The final classification loss function adopts a cross entropy loss function, and the rectangular box regression loss function adopts a SmoothL1 loss function.
S3: and detecting whether the image is tampered through the trained model.
And after the image to be detected is input into the trained image tampering detection model, the model outputs whether the image to be detected is a tampered image.
The method provided by the embodiment of the invention is beneficial to solving the problem that the conventional deep learning method is insufficient in extracting the forged features of the PS tampered image, and combines the RGB features of the forged region and the features of the resampling trace in the target detection network, thereby effectively improving the tampering detection and positioning capabilities of the PS image.
Example two:
the invention further provides an image tampering detection terminal device, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to realize the steps of the method embodiment of the first embodiment of the invention.
Further, as an executable scheme, the image tampering detection terminal device may be a desktop computer, a notebook, a palm computer, a cloud server, and other computing devices. The image tampering detection terminal device can include, but is not limited to, a processor and a memory. It is understood by those skilled in the art that the above-mentioned constituent structure of the image tampering detection terminal device is only an example of the image tampering detection terminal device, and does not constitute a limitation to the image tampering detection terminal device, and may include more or less components than the above, or combine some components, or different components, for example, the image tampering detection terminal device may further include an input/output device, a network access device, a bus, and the like, which is not limited in this embodiment of the present invention.
Further, as an executable solution, the Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, a discrete hardware component, and the like. The general-purpose processor may be a microprocessor or the processor may be any conventional processor, and the processor is a control center of the image tampering detection terminal device, and various interfaces and lines are used to connect various parts of the entire image tampering detection terminal device.
The memory may be configured to store the computer program and/or the module, and the processor may implement various functions of the image tamper detection terminal device by executing or executing the computer program and/or the module stored in the memory and calling data stored in the memory. The memory can mainly comprise a program storage area and a data storage area, wherein the program storage area can store an operating system and an application program required by at least one function; the storage data area may store data created according to the use of the mobile phone, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
The invention also provides a computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, carries out the steps of the above-mentioned method of an embodiment of the invention.
The module/unit integrated with the image tampering detection terminal device may be stored in a computer-readable storage medium if it is implemented in the form of a software function unit and sold or used as a separate product. Based on such understanding, all or part of the flow of the method according to the embodiments of the present invention may also be implemented by a computer program, which may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method embodiments described above may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), software distribution medium, and the like.
While the invention has been particularly shown and described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (7)
1. An image tampering detection method, characterized by comprising the steps of:
s1: acquiring images of two types, namely tampered images and images of non-tampered images to form a training set, and labeling tampered areas in the tampered images in the training set;
s2: constructing an image tampering detection model, and training the model through a training set;
a trunk part in a network structure of the image tampering detection model adopts a fast-RCNN network, after an input image is input into the fast-RCNN network, a candidate frame output by an RPN layer in the fast-RCNN network is input into an ROI posing layer in the fast-RCNN network on one hand, and a new ROI posing layer is input together with the input image on the other hand, an image output by the ROI posing layer is resampled to obtain a resampled spectrogram, the resampled spectrogram is input into an LSTM network to obtain a resampled characteristic diagram, the output of the resampled characteristic diagram after being subjected to an upsampling layer and a convolutional layer is input into a bilinear pooling layer together with the output of the fast-RCNN network to be fused, and the fused result is subjected to a full connection layer and a softmax layer to obtain a prediction result of whether the image is tampered or not;
s3: and detecting whether the image is tampered through the trained model.
2. The image tampering detection method according to claim 1, characterized in that: the resampling process comprises the following steps: and processing the candidate frame region in the image through an EM algorithm, and then performing Fourier transform to obtain a resample spectrogram.
3. The image tampering detection method according to claim 1, characterized in that: loss function L of image tampering detection modeltotalThe calculation formula of (2) is as follows:
Ltotal＝LRPN+Ltamper+Lbbox
wherein L isRPNRepresenting the loss function, L, of the RPN layertamperRepresents the final classification loss function, LbboxA rectangular box regression loss function is represented.
4. The image tampering detection method according to claim 3, characterized in that: the loss function of the RPN layer includes a classification loss function LclsAnd a regression loss function LregThe formula for calculating the loss function of the RPN layer is:
wherein, giRepresenting the predicted classification probability of the candidate box,
true tag representing candidate box, fiThe coordinates representing the prediction of the candidate box,
representing the real coordinates of the labeled frame corresponding to the candidate frame, i representing the serial number of the candidate frame, λ representing the weight coefficient, NclsIndicating the size of the small batch, NregIndicating the number of candidate boxes.
5. The image tampering detection method according to claim 3, characterized in that: the final classification loss function adopts a cross entropy loss function, and the rectangular box regression loss function adopts a SmoothL1 loss function.
6. An image tampering detection terminal device, characterized in that: comprising a processor, a memory and a computer program stored in the memory and running on the processor, the processor implementing the steps of the method according to any of claims 1 to 5 when executing the computer program.
7. A computer-readable storage medium storing a computer program, characterized in that: the computer program when executed by a processor implements the steps of the method as claimed in any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111533371.2ACN114444565B (en) | 2021-12-15 | 2021-12-15 | Image tampering detection method, terminal equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111533371.2ACN114444565B (en) | 2021-12-15 | 2021-12-15 | Image tampering detection method, terminal equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114444565Atrue CN114444565A (en) | 2022-05-06 |
| CN114444565B CN114444565B (en) | 2024-09-24 |
Family
ID=81363645
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111533371.2AActiveCN114444565B (en) | 2021-12-15 | 2021-12-15 | Image tampering detection method, terminal equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114444565B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115661725A (en)* | 2022-12-26 | 2023-01-31 | 浙江鹏信信息科技股份有限公司 | Method and system for detecting Deepfake video and readable storage medium |
| CN116758404A (en)* | 2023-05-12 | 2023-09-15 | 北京长木谷医疗科技股份有限公司 | Methods and devices for intelligently detecting the accuracy of medical images |
| CN118135641A (en)* | 2024-05-07 | 2024-06-04 | 齐鲁工业大学(山东省科学院) | Face counterfeiting detection method based on local counterfeiting area detection |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180253622A1 (en)* | 2017-03-06 | 2018-09-06 | Honda Motor Co., Ltd. | Systems for performing semantic segmentation and methods thereof |
| CN110349136A (en)* | 2019-06-28 | 2019-10-18 | 厦门大学 | A kind of tampered image detection method based on deep learning |
| CN110992238A (en)* | 2019-12-06 | 2020-04-10 | 上海电力大学 | A Blind Detection Method for Digital Image Tampering Based on Dual-Channel Network |
| CN111311563A (en)* | 2020-02-10 | 2020-06-19 | 北京工业大学 | Image tampering detection method based on multi-domain feature fusion |
| CN111553916A (en)* | 2020-05-09 | 2020-08-18 | 杭州中科睿鉴科技有限公司 | Image tampering area detection method based on multiple characteristics and convolutional neural network |
| CN112381775A (en)* | 2020-11-06 | 2021-02-19 | 厦门市美亚柏科信息股份有限公司 | Image tampering detection method, terminal device and storage medium |
- 2021
- 2021-12-15CNCN202111533371.2Apatent/CN114444565B/enactiveActive
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180253622A1 (en)* | 2017-03-06 | 2018-09-06 | Honda Motor Co., Ltd. | Systems for performing semantic segmentation and methods thereof |
| CN110349136A (en)* | 2019-06-28 | 2019-10-18 | 厦门大学 | A kind of tampered image detection method based on deep learning |
| CN110992238A (en)* | 2019-12-06 | 2020-04-10 | 上海电力大学 | A Blind Detection Method for Digital Image Tampering Based on Dual-Channel Network |
| CN111311563A (en)* | 2020-02-10 | 2020-06-19 | 北京工业大学 | Image tampering detection method based on multi-domain feature fusion |
| CN111553916A (en)* | 2020-05-09 | 2020-08-18 | 杭州中科睿鉴科技有限公司 | Image tampering area detection method based on multiple characteristics and convolutional neural network |
| CN112381775A (en)* | 2020-11-06 | 2021-02-19 | 厦门市美亚柏科信息股份有限公司 | Image tampering detection method, terminal device and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| P.ZHOU， ET AL,: "Learning Rich Features for Image Manipulation Detection", 《2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION》, 31 December 2018 (2018-12-31), pages 1053 - 1060* |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115661725A (en)* | 2022-12-26 | 2023-01-31 | 浙江鹏信信息科技股份有限公司 | Method and system for detecting Deepfake video and readable storage medium |
| CN116758404A (en)* | 2023-05-12 | 2023-09-15 | 北京长木谷医疗科技股份有限公司 | Methods and devices for intelligently detecting the accuracy of medical images |
| CN116758404B (en)* | 2023-05-12 | 2024-08-16 | 北京长木谷医疗科技股份有限公司 | Method and device for intelligently detecting accuracy of medical images |
| CN118135641A (en)* | 2024-05-07 | 2024-06-04 | 齐鲁工业大学(山东省科学院) | Face counterfeiting detection method based on local counterfeiting area detection |
| CN118135641B (en)* | 2024-05-07 | 2024-08-13 | 齐鲁工业大学(山东省科学院) | Face forgery detection method based on local forgery area detection |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114444565B (en) | 2024-09-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jiang et al. | Robust feature matching for remote sensing image registration via linear adaptive filtering | |
| CN108229490B (en) | Key point detection method, neural network training method, device and electronic equipment | |
| Hou et al. | BSNet: Dynamic hybrid gradient convolution based boundary-sensitive network for remote sensing image segmentation | |
| Liu et al. | Locating splicing forgery by fully convolutional networks and conditional random field | |
| CN114444565B (en) | Image tampering detection method, terminal equipment and storage medium | |
| CN111680690B (en) | Character recognition method and device | |
| CN112132145B (en) | An image classification method and system based on a model-extended convolutional neural network | |
| CN114049499A (en) | Target object detection method, device and storage medium for continuous contour | |
| CN107977661A (en) | The region of interest area detecting method decomposed based on full convolutional neural networks and low-rank sparse | |
| CN114581928B (en) | A table recognition method and system | |
| CN113449784A (en) | Image multi-classification method, device, equipment and medium based on prior attribute map | |
| CN115909378A (en) | Training method of receipt text detection model and receipt text detection method | |
| CN117218481A (en) | Fish identification method, device, equipment and storage medium | |
| WO2024174726A1 (en) | Handwritten and printed text detection method and device based on deep learning | |
| CN118429981A (en) | Smart electric meter registration identification method and device based on improvement YOLOv5 | |
| Košcevic et al. | Automatic visual reading of meters using deep learning | |
| CN113011506A (en) | Texture image classification method based on depth re-fractal spectrum network | |
| JP6151908B2 (en) | Learning device, identification device, and program thereof | |
| CN115937205A (en) | Surface defect tile image generation method, device, equipment and storage medium | |
| CN111488811A (en) | Face recognition method and device, terminal equipment and computer readable medium | |
| CN115272527A (en) | Image coloring method based on color disc countermeasure network | |
| CN112241736A (en) | Text detection method and device | |
| CN114511702A (en) | Remote sensing image segmentation method and system based on multi-scale weighted attention | |
| CN110020986A (en) | The single-frame image super-resolution reconstruction method remapped based on Euclidean subspace group two | |
| CN111667499A (en) | Image segmentation method, device and equipment for traffic signal lamp and storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |