CN112862700A - Noise attention-based blind denoising method for hyperspectral remote sensing image - Google Patents
Noise attention-based blind denoising method for hyperspectral remote sensing image Download PDFInfo
- Publication number
- CN112862700A CN112862700A CN202110058699.7A CN202110058699A CN112862700A CN 112862700 A CN112862700 A CN 112862700A CN 202110058699 A CN202110058699 A CN 202110058699A CN 112862700 A CN112862700 A CN 112862700A
- Authority
- CN
- China
- Prior art keywords
- denoising
- noise
- network
- intensity estimation
- noise intensity
- 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
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000000605 extraction Methods 0.000 claims description 13
- 230000001629 suppression Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 13
- 238000004422 calculation algorithm Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 238000013135 deep learning Methods 0.000 description 5
- 238000011084 recovery Methods 0.000 description 5
- 238000002474 experimental method Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 238000011160 research Methods 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000013256 Gubra-Amylin NASH model Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000003702 image correction Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
- G06T2207/10036—Multispectral image; Hyperspectral image
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a blind denoising method for hyperspectral remote sensing images based on noise attention, which acquires the noise intensity of different wave bands by constructing a noise estimation network, and inhibits the high noise wave band and enhances the low noise wave band by utilizing an attention mechanism; and then, a denoising network is constructed, useless information in the data is removed as much as possible, useful information is obtained, the denoising effect is improved, and the accuracy and the applicability of blind denoising are improved.
Description
Technical Field
The invention belongs to the technical field of image processing, and particularly relates to a blind denoising method for a hyperspectral remote sensing image.
Background
The hyperspectral remote sensing images originate from the early 60 s of the 20 th century, are fusion products of various high-end scientific technologies such as information processing technology, space measurement technology and computer science technology, and are widely applied to multiple fields such as military exploration, planetary detection, ground object classification and the like. Since the hyperspectral image contains a large amount of ground object information, people usually process and analyze the acquired hyperspectral data to acquire interesting and valuable information in order to extract more hyperspectral application values. In order to obtain the information, people often need high-quality hyperspectral images, so that the images can reflect the ground object information more truly as much as possible, and the result of further analysis is maximized. Unfortunately, almost all hyperspectral images are inevitably polluted by noise in the transmission process, the reason for the pollution is that equipment factors and external environment interference exist, and due to the doped useless interference noise, the subsequent research work on the images generates huge challenges and even the reliability of the processing technology is greatly reduced. Therefore, the research of popular and feasible hyperspectral image denoising methods is an academic target of a large number of scientific researchers, and the field of hyperspectral denoising also becomes an important research direction in the academic world.
In the past decades, a great many image recovery algorithms are proposed in succession, and the most common hyperspectral image denoising algorithm is to perform denoising processing on each band, and can be divided into a traditional algorithm and an algorithm based on deep learning. M.Magnetini et al, in the documents "M.Magnetini, V.Katkovnik, K.Egiazarian, non-local transform-domain filter for volumetric data denoising and reconstruction, in IEEE Transactions on Image Processing, 2013", propose a non-local self-similar denoising method, which adopts wavelet variation to perform threshold segmentation on the basis of three-dimensional block self-similarity, and then performs secondary denoising to obtain a satisfactory denoising effect. Zhang et al, in the literature "h.zhang, w.he, l.zhang, Hyperspectral image restoration using low-rank matrix recovery, in IEEE Transactions on geo science and remove Sensing,2014," proposed a denoising method based on low rank recovery, which directly learns images based on rank constraints, so that the effect of image recovery can be improved according to the constraints on the rank of the image matrix.
However, both the self-similarity method and the low rank matrix recovery method have limitations. Firstly, both have a large number of parameters manually set to control the quality of the denoising effect. Secondly, the algorithm has high calculation complexity and consumes a large amount of time for processing high-dimensional hyperspectral data.
In recent years, with the continuous development of computer science and technology, deep learning has made a significant breakthrough in the field of machine learning, and the deep learning has made a great effect improvement in the fields of image segmentation, image detection and recognition, image classification and the like, so that the deep learning is also applied in the field of hyperspectral image denoising to improve the image denoising effect. Chang et al in the document "Y.Chang, L.Yan, H.Fang, HSI-DeNet: Hyperspectral image restoration visual connected neural network, in IEEE Transactions on Geoscience and Remote Sensing, 2019" propose a simple 18-layer stacked neural network for denoising and restoring Hyperspectral images, and adopt a GAN model to perform global image correction, so as to obtain good restoration effect. On the basis of the documents, Q.Yuan, Q.Zhang, J.Li, et al, Hyperspectral image denoising a spatial-spectral residual denoising method, in IEEE Transactions on Geoscience and Remote Sensing,2019, proposes a multiscale feature extraction denoising method combining multispectral and single spectrum, improves the prior information supplement of an image by adding peripheral multispectral channel information when denoising a single channel, and further improves the denoising effect while denoising globally.
However, the above methods all have their own limitations. First, there are a lot of hyper-parameters in the traditional algorithm that need to be corrected manually, and for different data sets, a lot of parameter adjustment is often needed to obtain better effect. Secondly, the existing deep learning methods cannot well suppress non-uniform noise, and when the noise intensities of different wave bands are not uniform, the denoising robustness of the methods is deteriorated.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a hyperspectral remote sensing image blind denoising method based on noise attention, which acquires the noise intensity of different wave bands by constructing a noise estimation network, and inhibits the high-noise wave band by utilizing an attention mechanism, and enhances the low-noise wave band; and then, a denoising network is constructed, useless information in the data is removed as much as possible, useful information is obtained, the denoising effect is improved, and the accuracy and the applicability of blind denoising are improved.
The technical scheme adopted by the invention for solving the technical problem comprises the following steps:
step 1: constructing a noise intensity estimation network based on multilayer convolution;
the noise intensity estimation network is formed by directly connecting 15 3 multiplied by 3 convolutions in series; the input of the network is a noise image of H multiplied by W multiplied by C, the length and the width of the noise image are respectively H and W, and the number of channels is C; the output of the network is a noise intensity estimation graph of H multiplied by W multiplied by C, the length and the width of the noise intensity estimation graph are respectively H and W, and the number of channels is also C; the noise intensity estimation graph is used for estimating the noise intensity of each pixel in the noise image;
step 2: constructing a denoising network;
the Denoising network is formed by sequentially connecting four Denoising blocks and a Concat layer in series, wherein each Denoising Block comprises four multi-scale feature extraction units and a Concat layer;
the output of the first Denoising Block is sent to each subsequent Denoising Block and a Concat layer of a Denoising network; the outputs of the second and third Denoising blocks are sent to a Concat layer of a Denoising network besides the outputs of the subsequent Denoising blocks; the output of the fourth Denoising Block is sent to a Concat layer of the Denoising network;
the multi-scale feature extraction unit is formed by connecting a plurality of expansion convolutions; the features extracted by each multi-scale feature extraction unit are sent to a Concat layer in a Denoising Block and a subsequent multi-scale feature extraction unit;
and step 3: carrying out noise intensity estimation on a hyperspectral single wave band and K surrounding wave bands of the noise image by using a noise intensity estimation network to obtain a noise intensity estimation image of the estimated wave band;
and 4, step 4: performing high-noise suppression and low-noise enhancement on the noise intensity estimation graph of the estimated wave band obtained in the step (3) by using an attention mechanism;
and 5: inputting the noise intensity estimation graph of the estimated wave band processed in the step 4 into the denoising network constructed in the step 2 to remove noise;
step 6: and (3) converting the wave band of the noise intensity estimation in the step (3), and repeatedly executing the step (3) to the step (6) until the denoising of all the wave bands is finished, and finally achieving the blind denoising of the hyperspectral remote sensing data.
The invention has the following beneficial effects:
1. the denoising effect is better. According to the method, the noise intensity of different wave bands can be accurately obtained through the noise estimation attention network, the attention module is utilized to accurately inhibit and enhance the different wave bands, and therefore useless information in data is eliminated as far as possible, useful information is obtained, and the denoising effect is improved.
2. Blind denoising is more robust. According to the method, the noise prediction network is introduced, so that respective band noise levels can be obtained for any data set, the network can have a good blind denoising effect on different data sets in a self-adaptive manner, and the blind denoising robustness is higher.
3. The denoising process is more efficient. The denoising network of the method has better noise suppression capability, and simultaneously has less model parameters and algorithm complexity, thereby improving the denoising quality and efficiency.
Drawings
FIG. 1 is a block flow diagram of the method of the present invention.
FIG. 2 is a block diagram of a denoising network according to the method of the present invention.
Detailed Description
The invention is further illustrated with reference to the following figures and examples.
As shown in fig. 1, a blind denoising method for hyperspectral remote sensing images based on noise attention includes the following steps:
step 1: constructing a noise intensity estimation network based on multilayer convolution;
the noise intensity estimation network is formed by directly connecting 15 3 multiplied by 3 convolutions in series; the input of the network is a noise image of H multiplied by W multiplied by C, the length and the width of the noise image are respectively H and W, and the number of channels is C; the output of the network is a noise intensity estimation graph of H multiplied by W multiplied by C, the length and the width of the noise intensity estimation graph are respectively H and W, and the number of channels is also C; the noise intensity estimation graph is used for estimating the noise intensity of each pixel in the noise image;
step 2: constructing a denoising network;
the Denoising network is formed by sequentially connecting four Denoising blocks and a Concat layer in series, wherein each Denoising Block comprises four multi-scale feature extraction units and a Concat layer;
the output of the first Denoising Block is sent to each subsequent Denoising Block and a Concat layer of a Denoising network; the outputs of the second and third Denoising blocks are sent to a Concat layer of a Denoising network besides the outputs of the subsequent Denoising blocks; the output of the fourth Denoising Block is sent to a Concat layer of the Denoising network;
the multi-scale feature extraction unit is formed by connecting a plurality of expansion convolutions; the features extracted by each multi-scale feature extraction unit are sent to a Concat layer in a Denoising Block and a subsequent multi-scale feature extraction unit;
and step 3: carrying out noise intensity estimation on a hyperspectral single wave band and K surrounding wave bands of the noise image by using a noise intensity estimation network to obtain a noise intensity estimation image of the estimated wave band;
and 4, step 4: performing high-noise suppression and low-noise enhancement on the noise intensity estimation graph of the estimated wave band obtained in the step (3) by using an attention mechanism;
and 5: inputting the noise intensity estimation graph of the estimated wave band processed in the step 4 into the denoising network constructed in the step 2 to remove noise;
step 6: and (3) converting the wave band of the noise intensity estimation in the step (3), and repeatedly executing the step (3) to the step (6) until the denoising of all the wave bands is finished, and finally achieving the blind denoising of the hyperspectral remote sensing data.
The specific embodiment is as follows:
1. simulation conditions
The present embodiment is implemented by a central processing unitAnd (3) performing simulation on an i 5-34703.2 GHz CPU and a memory 16G, WINDOWS10 operating system by using MATLAB software.
The data set used in the experiment was fromWashington DC Mal of institutionsThe website of the remote sensing data set is https:// engineering. purdue. edu/. about. biehl/MultiSpec/hyperspectral. html. the data set comprises hyperspectral remote sensing data with 1280 length, 303 width and 191 wave bands, and experiments are carried out on the basis of the hyperspectral remote sensing data.
2. Emulated content
A clustering experiment was performed on the data set. In order to compare the effectiveness of the present invention, M.Magginoi et al, in the documents "M.Magginoi, V.Katkovnik, K.Egizarian, Nonlocal transform-domain filter for volumetric data denoising and reconstruction," in IEEE Transactions on Image Processing,2013, "mentioned non-local self-similar denoising method (BM4D)," H.Zhang et al, in "H.Zhang, W.Zhang, L.Zhang, Hyperspectral Image reconstruction using low-random matrix analysis, in IEEE Transactions on geographic classification and replacement, 2014," proposed rank in "proposed low-noise denoising Method (MR), and Q.Youin et al, on this basis, in the documents" Q.Q.S. transform, Q.S. transform, C.S. transform, N.S. transform and III.S. transform, found in the documents "average noise reduction and reconstruction" for comparing the average noise reduction and reconstruction with the average noise reduction method (I.S. transform, Q.S.S. transform on noise ratio, S.S. transform, N.S. transform Indexes such as Mean Structure Similarity (MSSIM) and the like are used for evaluating the quality of denoising in the experiment. The comparison results are shown in table 1 (highest values are indicated in bold).
Table 1 comparison of experimental results (σ ═ 10-70)
Method | LRMR | LRTV | BM4D | HSID-CNN | HSI-SDeCNN | Method for producing a composite material |
MPSNR | 30.88 | 30.64 | 27.59 | 30.63 | 30.45 | 32.86 |
MSSIM | 0.970 | 0.971 | 0.933 | 0.966 | 0.966 | 32.98 |
As can be seen from Table 1, the blind denoising performance of the method is superior to that of other comparison methods. The effectiveness of the invention can be verified through the simulation experiment.
Claims (1)
1. A blind denoising method for hyperspectral remote sensing images based on noise attention is characterized by comprising the following steps:
step 1: constructing a noise intensity estimation network based on multilayer convolution;
the noise intensity estimation network is formed by directly connecting 15 3 multiplied by 3 convolutions in series; the input of the network is a noise image of H multiplied by W multiplied by C, the length and the width of the noise image are respectively H and W, and the number of channels is C; the output of the network is a noise intensity estimation graph of H multiplied by W multiplied by C, the length and the width of the noise intensity estimation graph are respectively H and W, and the number of channels is also C; the noise intensity estimation graph is used for estimating the noise intensity of each pixel in the noise image;
step 2: constructing a denoising network;
the Denoising network is formed by sequentially connecting four Denoising blocks and a Concat layer in series, wherein each Denoising Block comprises four multi-scale feature extraction units and a Concat layer;
the output of the first Denoising Block is sent to each subsequent Denoising Block and a Concat layer of a Denoising network; the outputs of the second and third Denoising blocks are sent to a Concat layer of a Denoising network besides the outputs of the subsequent Denoising blocks; the output of the fourth Denoising Block is sent to a Concat layer of the Denoising network;
the multi-scale feature extraction unit is formed by connecting a plurality of expansion convolutions; the features extracted by each multi-scale feature extraction unit are sent to a Concat layer in a Denoising Block and a subsequent multi-scale feature extraction unit;
and step 3: carrying out noise intensity estimation on a hyperspectral single wave band and K surrounding wave bands of the noise image by using a noise intensity estimation network to obtain a noise intensity estimation image of the estimated wave band;
and 4, step 4: performing high-noise suppression and low-noise enhancement on the noise intensity estimation graph of the estimated wave band obtained in the step (3) by using an attention mechanism;
and 5: inputting the noise intensity estimation graph of the estimated wave band processed in the step 4 into the denoising network constructed in the step 2 to remove noise;
step 6: and (3) converting the wave band of the noise intensity estimation in the step (3), and repeatedly executing the step (3) to the step (6) until the denoising of all the wave bands is finished, and finally achieving the blind denoising of the hyperspectral remote sensing data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110058699.7A CN112862700B (en) | 2021-01-16 | 2021-01-16 | Hyperspectral remote sensing image blind denoising method based on noise attention |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110058699.7A CN112862700B (en) | 2021-01-16 | 2021-01-16 | Hyperspectral remote sensing image blind denoising method based on noise attention |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112862700A true CN112862700A (en) | 2021-05-28 |
CN112862700B CN112862700B (en) | 2023-10-27 |
Family
ID=76005881
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110058699.7A Active CN112862700B (en) | 2021-01-16 | 2021-01-16 | Hyperspectral remote sensing image blind denoising method based on noise attention |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112862700B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113436111A (en) * | 2021-07-21 | 2021-09-24 | 西北工业大学 | Hyperspectral remote sensing image denoising method based on network structure search |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20140145801A (en) * | 2013-06-14 | 2014-12-24 | 전북대학교산학협력단 | System and method for improving quality in images using neural network |
CN109658348A (en) * | 2018-11-16 | 2019-04-19 | 天津大学 | The estimation of joint noise and image de-noising method based on deep learning |
CN110516728A (en) * | 2019-08-20 | 2019-11-29 | 西安电子科技大学 | Polarization SAR terrain classification method based on denoising convolutional neural networks |
CN110570364A (en) * | 2019-08-05 | 2019-12-13 | 天津大学 | depth neural network-based sub-focal plane polarization image denoising method |
CN111260591A (en) * | 2020-03-12 | 2020-06-09 | 武汉大学 | Image self-adaptive denoising method based on attention mechanism |
CN111292259A (en) * | 2020-01-14 | 2020-06-16 | 西安交通大学 | Deep learning image denoising method integrating multi-scale and attention mechanism |
US20200234402A1 (en) * | 2019-01-18 | 2020-07-23 | Ramot At Tel-Aviv University Ltd. | Method and system for end-to-end image processing |
CN111931738A (en) * | 2020-09-28 | 2020-11-13 | 南京邮电大学 | Neural network model pre-training method and device for remote sensing image |
CN112233026A (en) * | 2020-09-29 | 2021-01-15 | 南京理工大学 | SAR image denoising method based on multi-scale residual attention network |
-
2021
- 2021-01-16 CN CN202110058699.7A patent/CN112862700B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20140145801A (en) * | 2013-06-14 | 2014-12-24 | 전북대학교산학협력단 | System and method for improving quality in images using neural network |
CN109658348A (en) * | 2018-11-16 | 2019-04-19 | 天津大学 | The estimation of joint noise and image de-noising method based on deep learning |
US20200234402A1 (en) * | 2019-01-18 | 2020-07-23 | Ramot At Tel-Aviv University Ltd. | Method and system for end-to-end image processing |
CN110570364A (en) * | 2019-08-05 | 2019-12-13 | 天津大学 | depth neural network-based sub-focal plane polarization image denoising method |
CN110516728A (en) * | 2019-08-20 | 2019-11-29 | 西安电子科技大学 | Polarization SAR terrain classification method based on denoising convolutional neural networks |
CN111292259A (en) * | 2020-01-14 | 2020-06-16 | 西安交通大学 | Deep learning image denoising method integrating multi-scale and attention mechanism |
CN111260591A (en) * | 2020-03-12 | 2020-06-09 | 武汉大学 | Image self-adaptive denoising method based on attention mechanism |
CN111931738A (en) * | 2020-09-28 | 2020-11-13 | 南京邮电大学 | Neural network model pre-training method and device for remote sensing image |
CN112233026A (en) * | 2020-09-29 | 2021-01-15 | 南京理工大学 | SAR image denoising method based on multi-scale residual attention network |
Non-Patent Citations (1)
Title |
---|
张静妙;高双喜;王晓娜;: "基于低秩字典学习的高光谱遥感图像去噪", 控制工程, no. 06, pages 823 - 827 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113436111A (en) * | 2021-07-21 | 2021-09-24 | 西北工业大学 | Hyperspectral remote sensing image denoising method based on network structure search |
CN113436111B (en) * | 2021-07-21 | 2024-01-09 | 西北工业大学 | Hyperspectral remote sensing image denoising method based on network structure search |
Also Published As
Publication number | Publication date |
---|---|
CN112862700B (en) | 2023-10-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Tian et al. | Attention-guided CNN for image denoising | |
Chen et al. | Denoising hyperspectral image with non-iid noise structure | |
Liu et al. | A 3-D atrous convolution neural network for hyperspectral image denoising | |
Doegar et al. | Cnn based image forgery detection using pre-trained alexnet model | |
CN112233026A (en) | SAR image denoising method based on multi-scale residual attention network | |
Khmag et al. | Clustering-based natural image denoising using dictionary learning approach in wavelet domain | |
Fan et al. | Superpixel guided deep-sparse-representation learning for hyperspectral image classification | |
CN110598584A (en) | Convolutional neural network face recognition algorithm based on wavelet transform and DCT | |
CN111339924B (en) | Polarized SAR image classification method based on superpixel and full convolution network | |
Liu et al. | True wide convolutional neural network for image denoising | |
Dhar et al. | Accurate segmentation of complex document image using digital shearlet transform with neutrosophic set as uncertainty handling tool | |
CN112862700A (en) | Noise attention-based blind denoising method for hyperspectral remote sensing image | |
CN112613583B (en) | High-frequency information extraction clustering method for low-frequency noise face image | |
Arya et al. | ADMM optimizer for integrating wavelet-patch and group-based sparse representation for image inpainting | |
CN116310452B (en) | Multi-view clustering method and system | |
Zhong et al. | Multiscale feature extraction based on convolutional sparse decomposition for hyperspectral image classification | |
Wu et al. | Dcanet: Dual convolutional neural network with attention for image blind denoising | |
Recoskie et al. | Learning filters for the 2D wavelet transform | |
CN111127407A (en) | Fourier transform-based style migration counterfeit image detection device and method | |
Geetha et al. | Savitzky golay and kpca based optimal discrete wavelet transform architecture for image compression | |
CN113436111A (en) | Hyperspectral remote sensing image denoising method based on network structure search | |
Gattim et al. | ROTATION AND SCALE INVARIANT FEATURE EXTRACTION FOR MRI BRAIN IMAGES. | |
Xu et al. | An efficient method for human face recognition using nonsubsampled contourlet transform and support vector machine | |
Tojo et al. | Image Denoising Using Multi Scaling Aided Double Decker Convolutional Neural Network | |
Chen et al. | Facial Age Estimation based on Discrete Wavelet Transform-Deep Convolutional Neural Networks |
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 |