CN111783545A - Coal mine unmanned aerial vehicle image enhancement method based on generation of countermeasure network - Google Patents
Coal mine unmanned aerial vehicle image enhancement method based on generation of countermeasure network Download PDFInfo
- Publication number
- CN111783545A CN111783545A CN202010490445.8A CN202010490445A CN111783545A CN 111783545 A CN111783545 A CN 111783545A CN 202010490445 A CN202010490445 A CN 202010490445A CN 111783545 A CN111783545 A CN 111783545A
- Authority
- CN
- China
- Prior art keywords
- image
- generator
- discriminator
- aerial vehicle
- unmanned aerial
- 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
Links
- 239000003245 coal Substances 0.000 title claims abstract description 34
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000012549 training Methods 0.000 claims abstract description 41
- 238000012545 processing Methods 0.000 claims abstract description 23
- 230000004913 activation Effects 0.000 claims description 12
- 238000013507 mapping Methods 0.000 claims description 11
- 238000010606 normalization Methods 0.000 claims description 8
- 238000013527 convolutional neural network Methods 0.000 claims description 6
- 238000007689 inspection Methods 0.000 abstract description 8
- 230000006870 function Effects 0.000 description 31
- 238000010586 diagram Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000012886 linear function Methods 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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; CALCULATING OR 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the technical field of image processing, in particular to a coal mine unmanned aerial vehicle image enhancement method based on a generation countermeasure network. It includes: s1, acquiring a clear image shot by the coal mine unmanned aerial vehicle, processing the acquired image and further constructing a training set; step S2, constructing a generating countermeasure network, wherein the constructed generating countermeasure network comprises a generator and a discriminator; step S3, training the generator by fixing the parameters of the discriminator; step S4, training the discriminator by fixing the parameters of the generator; step S5, repeating the steps S3 and S4, and updating the parameters of the generator and the discriminator according to the loss function until the image generated by the generator is discriminated to be a clear image by the discriminator, and ending the model training; and step S6, processing the blurred image shot by the unmanned aerial vehicle, inputting the processed blurred image into a trained generator, and taking the corresponding image output by the generator as an enhanced sharp image. The invention can better realize the enhancement processing of the inspection image of the unmanned aerial vehicle.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a coal mine unmanned aerial vehicle image enhancement method based on a generation countermeasure network.
Background
In recent years, along with the rapid development of the unmanned aerial vehicle technology, a plurality of applications of the unmanned aerial vehicle inspection technology appear, and particularly in the field of coal mines, the manual inspection mode has low efficiency and large workload, and has great potential safety hazards due to the limitation of coal mine regions. Unmanned aerial vehicle patrols and examines through the visible light camera, the infrared camera of carrying on, laser radar range finding sensor etc. and replace the line patrol and examine the work, realize functions such as target detection and target identification. However, due to the complex environmental factors of coal mine regions, the images shot by the unmanned aerial vehicle have the problems of serious noise, loss of details and the like, and the small targets in the images are difficult to effectively detect and identify. Therefore, the image enhancement of the coal mine unmanned aerial vehicle has great significance for the wide application of the unmanned aerial vehicle inspection technology.
The traditional image enhancement methods mainly increase the contrast of an image through histogram equalization, laplacian transform and the like, but the methods do not consider context information in the image and cannot obtain a real and clear image.
Disclosure of Invention
The invention provides a coal mine unmanned aerial vehicle image enhancement method based on a generation countermeasure network, which can overcome certain defects in the prior art.
The invention discloses a coal mine unmanned aerial vehicle image enhancement method based on a generation countermeasure network, which comprises the following steps:
step S1, acquiring a clear image shot by the coal mine unmanned aerial vehicle and processing the acquired image to further construct a training set, wherein a single element of the training set is a binary group, and the binary group comprises the clear image and a fuzzy image corresponding to the clear image;
step S2, constructing a generating countermeasure network, wherein the constructed generating countermeasure network comprises a generator and a discriminator;
step S3, fixing the parameters of the discriminator, inputting the fuzzy image in the training set into the generator, and discriminating whether the image generated by the generator is clear through the discriminator to realize the training of the generator;
step S4, fixing the parameters of the generator, and inputting the clear images in the training set and the images generated by the generator into the discriminator respectively to realize the training of the discriminator;
step S5, repeating the step S3 and the step S4, and updating the parameters of the generator and the discriminator according to the loss function until the image generated by the generator is discriminated to be a clear image by the discriminator, and finishing the model training;
and step S6, processing the blurred image shot by the unmanned aerial vehicle, inputting the processed blurred image into a trained generator, and taking the corresponding image output by the generator as an enhanced sharp image.
Preferably, in step S1, the processing performed on the acquired image includes the steps of,
step S11, acquiring a blurred image corresponding to the corresponding sharp image by adding noise according to the acquired image;
step S12, the pixel sizes of the corresponding sharp image and the corresponding blurred image are set, thereby forming a single element of the training set.
Preferably, in step S2, the generator uses a convolutional neural network, which includes a linear mapping layer and four convolutional layers, the convolutional kernel of each convolutional layer has a size of 5 × 5, the linear mapping layer and the first three convolutional layers are respectively connected with the batch normalization and the ReLU activation function, and the last convolutional layer is connected with the Tanh activation function.
Preferably, the discriminator adopts a convolutional neural network, which comprises four convolutional layers and a linear mapping layer, the size of the convolutional core of each convolutional layer is 5 × 5, the first convolutional layer is connected with an LReLU activation function, and the three subsequent convolutional layers are respectively connected with batch normalization and LReLU activation functions.
Preferably, in step S3, an error back propagation algorithm is used to fix the parameters of the arbiter and minimize the loss function of the generator.
Preferably, in step S4, an error back propagation algorithm is used to fix the generator parameters and minimize the loss function of the arbiter.
Preferably, the loss function of the generator is
Wherein G is the generator model, D is the discriminator model, z is random noise, and x is the blurred image.
Preferably, the penalty function of the discriminator is
Wherein G is the generator model, D is the discriminator model, x is the blurred image, and y is the sharp image.
The method of the invention is characterized in that:
1. the target image of the polling can be acquired by the coal mine unmanned aerial vehicle, the target image is preprocessed, and the preprocessed image is input into the generation countermeasure network for processing, so that the image acquired by the coal mine unmanned aerial vehicle can be better enhanced by the generation countermeasure network, the quality of a fuzzy image in a polling scene of the coal mine unmanned aerial vehicle can be effectively improved, and the details of the generated image are richer;
2. the generation countermeasure network is trained by establishing a training set, and parameters of a generator and a discriminator can be preferably determined, so that new images acquired by the coal mine unmanned aerial vehicle through routing inspection can be preferably enhanced;
3. by introducing a mean square error penalty and a gradient penalty function in the training process of generating the countermeasure network, the generated enhanced image with richer details can be generated aiming at the problems of low brightness, low contrast, image blurring and the like of the shot image when the model after final training is used for processing a newly acquired image.
Drawings
Fig. 1 is a schematic flow chart of a coal mine unmanned aerial vehicle image enhancement method in embodiment 1;
FIG. 2 is a schematic diagram of a model for generating a countermeasure network in example 1;
FIG. 3 is a model diagram of a generator in embodiment 1;
FIG. 4 is a schematic model diagram of an arbiter in example 1;
fig. 5 is an effect diagram of the coal mine unmanned aerial vehicle image enhancement method in embodiment 1.
Detailed Description
For a further understanding of the invention, reference should be made to the following detailed description taken in conjunction with the accompanying drawings and examples. It is to be understood that the examples are illustrative of the invention and not limiting.
Example 1
With reference to fig. 1 and 2, the embodiment provides a coal mine unmanned aerial vehicle image enhancement method based on generation of a countermeasure network, which includes the following steps:
step S1, acquiring a clear image shot by the coal mine unmanned aerial vehicle and processing the acquired image to further construct a training set, wherein a single element of the training set is a binary group, and the binary group comprises the clear image and a fuzzy image corresponding to the clear image;
step S2, constructing a generating countermeasure network, wherein the constructed generating countermeasure network comprises a generator and a discriminator;
step S3, fixing the parameters of the discriminator, inputting the fuzzy image in the training set into the generator, and discriminating whether the image generated by the generator is clear through the discriminator to realize the training of the generator;
s4, fixing the parameters of the generator, and respectively inputting the clear images in the training set and the images generated by the generator into the discriminator to realize the training of the discriminator;
step S5, repeating the step S3 and the step S4, and updating the parameters of the generator and the discriminator according to the loss function until the image generated by the generator is discriminated to be a clear image by the discriminator, and finishing the model training;
and step S6, processing the blurred image shot by the unmanned aerial vehicle, inputting the processed blurred image into a trained generator, and taking the corresponding image output by the generator as an enhanced sharp image.
By the method in the embodiment, the target image of the routing inspection can be acquired through the coal mine unmanned aerial vehicle, the target image is preprocessed, and the preprocessed image is input into the generation countermeasure network for processing, so that the image acquired by the coal mine unmanned aerial vehicle can be better enhanced through the generation countermeasure network, the fuzzy image quality of the coal mine unmanned aerial vehicle in the routing inspection scene can be effectively improved, and the generated image details are richer.
The generation countermeasure network comprises a generator and a discriminator, wherein the generator is used for generating a new image from the random noise image and the fuzzy image, and the discriminator is used for distinguishing the generated image from the clear image. The generation countermeasure network is trained by establishing a training set, parameters of the generator and the discriminator can be preferably determined, and therefore the enhancement processing of new images acquired by the coal mine unmanned aerial vehicle through patrol inspection can be preferably realized.
In this embodiment, the training set includes 10000 clear pictures that are shot by the unmanned aerial vehicle in the coal mine and add the fuzzy picture that the noise was acquireed according to this 10000 clear pictures.
In step S1 of the present embodiment, the processing performed on the acquired image includes the steps of,
step S11, acquiring a blurred image corresponding to the corresponding clear image in a noise adding mode according to the acquired image;
and step S12, setting the pixel size of the corresponding sharp image and the corresponding blurred image, and further forming a single element of the training set.
In this embodiment, through the processing in step S11, the blurred image corresponding to the corresponding sharp image can be obtained by adding random noise to the obtained sharp image, so that the training set can be preferably obtained; by the processing in step S12, all images can be processed to the same pixel size, and therefore, the subsequent image processing can be facilitated.
In this embodiment, the pixel size of the picture is processed to 64 × 64 in step S12.
Referring to fig. 3, in step S2 of this embodiment, the generator uses a convolutional neural network, which includes a linear mapping layer and four convolutional layers, the size of the convolutional kernel of each convolutional layer is 5 × 5, the batch normalization and the ReLU activation functions are respectively connected to the back of the linear mapping layer and the first three convolutional layers, and the Tanh activation function is connected to the back of the last convolutional layer.
In the generator of this embodiment, a Linear mapping layer adopts a Linear function, a convolution layer adopts a Deconv function, and a batch normalization function is a BN function.
Referring to fig. 4, the discriminator uses a convolutional neural network, which includes four convolutional layers and a linear mapping layer, the size of the convolutional core of each convolutional layer is 5 × 5, the lreul activation function is connected to the first convolutional layer, and the batch normalization and the lreul activation function are connected to the three subsequent convolutional layers.
In the discriminator of the present embodiment, the Linear mapping layer employs a Linear function, the convolution layer employs a Conv function, and the batch normalization function is a BN function.
In step S3 of this embodiment, an error back propagation algorithm is used to fix the parameters of the discriminator and minimize the loss function of the generator. Thereby enabling a better training of the generator.
In step S4 of this embodiment, an error back propagation algorithm is used to fix the generator parameters and minimize the loss function of the discriminator. Thereby enabling better training of the discriminators.
In this embodiment, the loss function of the generator is
Wherein G is the generator model, D is the discriminator model, z is random noise, and x is the blurred image.
In this embodiment, the penalty function of the discriminator is
Wherein G is the generator model, D is the discriminator model, x is the blurred image, and y is the sharp image.
In this embodiment, a mean square error penalty is introduced into a loss function of the generator to make the generated image closer to a corresponding clear image, and the error penalty function is pG＝||G(z|x)-y||2. Wherein G is the generator model, z is random noise, x is a blurred image, and y is a sharp image.
In this embodiment, a gradient penalty is introduced into the loss function of the discriminator to stably generate the training process of the countermeasure network, and the gradient penalty function is
Wherein D is the discriminator model, x is the blurred image,
in order to generate the image(s),
α is [0,1 ] for the gradient value of the discriminator model]A random value in between.
In the embodiment, by introducing a mean square error penalty and a gradient penalty function in the training process of generating the countermeasure network, when a model after final training is used for processing a newly acquired image, the generated enhanced image with richer details can be generated aiming at the problems of low brightness, low contrast, image blurring and the like of a shot image.
In step S6 of the present embodiment, the processing of the blurred image captured by the drone is mainly to crop the image size to a predetermined pixel size, which in the present embodiment is 64 × 64 pixel size.
More specifically, the parameters for generating the countermeasure network of the present embodiment are set such that the batch size during training is 32, the convolution kernel size of the generator and the arbiter is 5 × 5, and the step size is 2. An Adam optimizer is adopted in the training process of the generator and the discriminator, the learning rate of the generator is 0.0001, the learning rate of the discriminator is 0.0005, and the dimension of noise is 100.
Fig. 5 is a diagram showing effects of the image enhancement method according to the present embodiment. The left image is an original blurred image shot by the unmanned aerial vehicle, and the right image is a clear image after the generation of the confrontation network image enhancement. The image enhancement method based on the generation countermeasure network has good effect on the aspects of image definition and reality.
The present invention and its embodiments have been described above schematically, without limitation, and what is shown in the drawings is only one of the embodiments of the present invention, and the actual structure is not limited thereto. Therefore, if the person skilled in the art receives the teaching, without departing from the spirit of the invention, the person skilled in the art shall not inventively design the similar structural modes and embodiments to the technical solution, but shall fall within the scope of the invention.
Claims (8)
1. A coal mine unmanned aerial vehicle image enhancement method based on generation of a countermeasure network comprises the following steps:
step S1, acquiring a clear image shot by the coal mine unmanned aerial vehicle and processing the acquired image to further construct a training set, wherein a single element of the training set is a binary group, and the binary group comprises the clear image and a fuzzy image corresponding to the clear image;
step S2, constructing a generating countermeasure network, wherein the constructed generating countermeasure network comprises a generator and a discriminator;
step S3, fixing the parameters of the discriminator, inputting the fuzzy image in the training set into the generator, and discriminating whether the image generated by the generator is clear through the discriminator to realize the training of the generator;
step S4, fixing the parameters of the generator, and inputting the clear images in the training set and the images generated by the generator into the discriminator respectively to realize the training of the discriminator;
step S5, repeating the step S3 and the step S4, and updating the parameters of the generator and the discriminator according to the loss function until the image generated by the generator is discriminated to be a clear image by the discriminator, and finishing the model training;
and step S6, processing the blurred image shot by the unmanned aerial vehicle, inputting the processed blurred image into a trained generator, and taking the corresponding image output by the generator as an enhanced sharp image.
2. The coal mine unmanned aerial vehicle image enhancement method based on the generation countermeasure network of claim 1, wherein: in step S1, the processing performed on the acquired image includes the steps of,
step S11, acquiring a blurred image corresponding to the corresponding sharp image by adding noise according to the acquired image;
step S12, the pixel sizes of the corresponding sharp image and the corresponding blurred image are set, thereby forming a single element of the training set.
3. The coal mine unmanned aerial vehicle image enhancement method based on the generation countermeasure network of claim 1, wherein: in step S2, the generator uses a convolutional neural network, which includes a linear mapping layer and four convolutional layers, the size of the convolutional core of each convolutional layer is 5 × 5, the back of the linear mapping layer and the first three convolutional layers is respectively connected to the batch normalization and the ReLU activation function, and the back of the last convolutional layer is connected to the Tanh activation function.
4. The coal mine unmanned aerial vehicle image enhancement method based on the generation countermeasure network of claim 1, wherein: the discriminator adopts a convolutional neural network and comprises four convolutional layers and a linear mapping layer, the size of a convolutional core of each convolutional layer is 5 multiplied by 5, an LReLU activation function is connected behind the first convolutional layer, and a batch normalization function and an LReLU activation function are respectively connected behind the three subsequent convolutional layers.
5. The coal mine unmanned aerial vehicle image enhancement method based on the generation countermeasure network of claim 1, wherein: in step S3, an error back propagation algorithm is used to fix the parameters of the discriminator and minimize the loss function of the generator.
6. The coal mine unmanned aerial vehicle image enhancement method based on the generation countermeasure network of claim 1, wherein: in step S4, an error back propagation algorithm is used to fix the generator parameters and minimize the loss function of the arbiter.
7. The coal mine unmanned aerial vehicle image enhancement method based on the generation countermeasure network of claim 1, wherein: the loss function of the generator is
Wherein G is the generator model, D is the discriminator model, z is random noise, and x is the blurred image.
8. The coal mine unmanned aerial vehicle image enhancement method based on the generation countermeasure network of claim 1, wherein: the loss function of the discriminator is
Wherein G is the generator model, D is the discriminator model, x is the blurred image, and y is the sharp image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010490445.8A CN111783545A (en) | 2020-06-02 | 2020-06-02 | Coal mine unmanned aerial vehicle image enhancement method based on generation of countermeasure network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010490445.8A CN111783545A (en) | 2020-06-02 | 2020-06-02 | Coal mine unmanned aerial vehicle image enhancement method based on generation of countermeasure network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111783545A true CN111783545A (en) | 2020-10-16 |
Family
ID=72753314
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010490445.8A Pending CN111783545A (en) | 2020-06-02 | 2020-06-02 | Coal mine unmanned aerial vehicle image enhancement method based on generation of countermeasure network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111783545A (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112668529A (en) * | 2020-12-31 | 2021-04-16 | 神思电子技术股份有限公司 | Dish sample image enhancement identification method |
| CN113487504A (en) * | 2021-07-01 | 2021-10-08 | 天津大学 | Underwater gray level polarization image restoration method based on generation countermeasure network |
| CN113534837A (en) * | 2021-07-06 | 2021-10-22 | 深圳市科卫泰实业发展有限公司 | Unmanned aerial vehicle flight state fusion control system and control method based on generation of countermeasure network |
| US11403069B2 (en) | 2017-07-24 | 2022-08-02 | Tesla, Inc. | Accelerated mathematical engine |
| US11409692B2 (en) | 2017-07-24 | 2022-08-09 | Tesla, Inc. | Vector computational unit |
| US11487288B2 (en) | 2017-03-23 | 2022-11-01 | Tesla, Inc. | Data synthesis for autonomous control systems |
| CN115310361A (en) * | 2022-08-16 | 2022-11-08 | 中国矿业大学 | Method and system for predicting underground dust concentration of coal mine based on WGAN-CNN |
| US11537811B2 (en) | 2018-12-04 | 2022-12-27 | Tesla, Inc. | Enhanced object detection for autonomous vehicles based on field view |
| US11562231B2 (en) | 2018-09-03 | 2023-01-24 | Tesla, Inc. | Neural networks for embedded devices |
| US11561791B2 (en) | 2018-02-01 | 2023-01-24 | Tesla, Inc. | Vector computational unit receiving data elements in parallel from a last row of a computational array |
| US11567514B2 (en) | 2019-02-11 | 2023-01-31 | Tesla, Inc. | Autonomous and user controlled vehicle summon to a target |
| US11610117B2 (en) | 2018-12-27 | 2023-03-21 | Tesla, Inc. | System and method for adapting a neural network model on a hardware platform |
| US11636333B2 (en) | 2018-07-26 | 2023-04-25 | Tesla, Inc. | Optimizing neural network structures for embedded systems |
| US11665108B2 (en) | 2018-10-25 | 2023-05-30 | Tesla, Inc. | QoS manager for system on a chip communications |
| US11681649B2 (en) | 2017-07-24 | 2023-06-20 | Tesla, Inc. | Computational array microprocessor system using non-consecutive data formatting |
| US11734562B2 (en) | 2018-06-20 | 2023-08-22 | Tesla, Inc. | Data pipeline and deep learning system for autonomous driving |
| US11748620B2 (en) | 2019-02-01 | 2023-09-05 | Tesla, Inc. | Generating ground truth for machine learning from time series elements |
| US11790664B2 (en) | 2019-02-19 | 2023-10-17 | Tesla, Inc. | Estimating object properties using visual image data |
| US11816585B2 (en) | 2018-12-03 | 2023-11-14 | Tesla, Inc. | Machine learning models operating at different frequencies for autonomous vehicles |
| US11841434B2 (en) | 2018-07-20 | 2023-12-12 | Tesla, Inc. | Annotation cross-labeling for autonomous control systems |
| US11893393B2 (en) | 2017-07-24 | 2024-02-06 | Tesla, Inc. | Computational array microprocessor system with hardware arbiter managing memory requests |
| US11893774B2 (en) | 2018-10-11 | 2024-02-06 | Tesla, Inc. | Systems and methods for training machine models with augmented data |
| US12014553B2 (en) | 2019-02-01 | 2024-06-18 | Tesla, Inc. | Predicting three-dimensional features for autonomous driving |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110264424A (en) * | 2019-06-20 | 2019-09-20 | 北京理工大学 | A kind of fuzzy retinal fundus images Enhancement Method based on generation confrontation network |
| WO2020103171A1 (en) * | 2018-11-21 | 2020-05-28 | 北京大学深圳研究生院 | Bi-level optimization method for image deblurring |
-
2020
- 2020-06-02 CN CN202010490445.8A patent/CN111783545A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020103171A1 (en) * | 2018-11-21 | 2020-05-28 | 北京大学深圳研究生院 | Bi-level optimization method for image deblurring |
| CN110264424A (en) * | 2019-06-20 | 2019-09-20 | 北京理工大学 | A kind of fuzzy retinal fundus images Enhancement Method based on generation confrontation network |
Non-Patent Citations (2)
| Title |
|---|
| ZHANG GUOBIN 等: ""A Classification method for adaptive relay protection setting system Based on clustering analysis"", 《PROCEEDINGS OF 2011 INTERNATIONAL CONFERENCE ON ADVANCED POWER SYSTEM AUTOMATION AND PROTECTION(APAP 2011)》, 16 October 2011 (2011-10-16) * |
| 裴慧坤 等: ""基于生成对抗网络的无人机图像去模糊方法"", 《地理空间信息》, vol. 17, no. 12, 31 December 2019 (2019-12-31), pages 4 - 8 * |
Cited By (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US12020476B2 (en) | 2017-03-23 | 2024-06-25 | Tesla, Inc. | Data synthesis for autonomous control systems |
| US11487288B2 (en) | 2017-03-23 | 2022-11-01 | Tesla, Inc. | Data synthesis for autonomous control systems |
| US11893393B2 (en) | 2017-07-24 | 2024-02-06 | Tesla, Inc. | Computational array microprocessor system with hardware arbiter managing memory requests |
| US11403069B2 (en) | 2017-07-24 | 2022-08-02 | Tesla, Inc. | Accelerated mathematical engine |
| US11409692B2 (en) | 2017-07-24 | 2022-08-09 | Tesla, Inc. | Vector computational unit |
| US11681649B2 (en) | 2017-07-24 | 2023-06-20 | Tesla, Inc. | Computational array microprocessor system using non-consecutive data formatting |
| US11561791B2 (en) | 2018-02-01 | 2023-01-24 | Tesla, Inc. | Vector computational unit receiving data elements in parallel from a last row of a computational array |
| US11797304B2 (en) | 2018-02-01 | 2023-10-24 | Tesla, Inc. | Instruction set architecture for a vector computational unit |
| US11734562B2 (en) | 2018-06-20 | 2023-08-22 | Tesla, Inc. | Data pipeline and deep learning system for autonomous driving |
| US11841434B2 (en) | 2018-07-20 | 2023-12-12 | Tesla, Inc. | Annotation cross-labeling for autonomous control systems |
| US11636333B2 (en) | 2018-07-26 | 2023-04-25 | Tesla, Inc. | Optimizing neural network structures for embedded systems |
| US11562231B2 (en) | 2018-09-03 | 2023-01-24 | Tesla, Inc. | Neural networks for embedded devices |
| US11983630B2 (en) | 2018-09-03 | 2024-05-14 | Tesla, Inc. | Neural networks for embedded devices |
| US11893774B2 (en) | 2018-10-11 | 2024-02-06 | Tesla, Inc. | Systems and methods for training machine models with augmented data |
| US11665108B2 (en) | 2018-10-25 | 2023-05-30 | Tesla, Inc. | QoS manager for system on a chip communications |
| US11816585B2 (en) | 2018-12-03 | 2023-11-14 | Tesla, Inc. | Machine learning models operating at different frequencies for autonomous vehicles |
| US11537811B2 (en) | 2018-12-04 | 2022-12-27 | Tesla, Inc. | Enhanced object detection for autonomous vehicles based on field view |
| US11908171B2 (en) | 2018-12-04 | 2024-02-20 | Tesla, Inc. | Enhanced object detection for autonomous vehicles based on field view |
| US11610117B2 (en) | 2018-12-27 | 2023-03-21 | Tesla, Inc. | System and method for adapting a neural network model on a hardware platform |
| US11748620B2 (en) | 2019-02-01 | 2023-09-05 | Tesla, Inc. | Generating ground truth for machine learning from time series elements |
| US12014553B2 (en) | 2019-02-01 | 2024-06-18 | Tesla, Inc. | Predicting three-dimensional features for autonomous driving |
| US11567514B2 (en) | 2019-02-11 | 2023-01-31 | Tesla, Inc. | Autonomous and user controlled vehicle summon to a target |
| US11790664B2 (en) | 2019-02-19 | 2023-10-17 | Tesla, Inc. | Estimating object properties using visual image data |
| CN112668529A (en) * | 2020-12-31 | 2021-04-16 | 神思电子技术股份有限公司 | Dish sample image enhancement identification method |
| CN113487504A (en) * | 2021-07-01 | 2021-10-08 | 天津大学 | Underwater gray level polarization image restoration method based on generation countermeasure network |
| CN113534837A (en) * | 2021-07-06 | 2021-10-22 | 深圳市科卫泰实业发展有限公司 | Unmanned aerial vehicle flight state fusion control system and control method based on generation of countermeasure network |
| CN113534837B (en) * | 2021-07-06 | 2024-03-22 | 深圳市科卫泰实业发展有限公司 | Unmanned aerial vehicle flight state fusion control system and control method based on generation countermeasure network |
| CN115310361A (en) * | 2022-08-16 | 2022-11-08 | 中国矿业大学 | Method and system for predicting underground dust concentration of coal mine based on WGAN-CNN |
| CN115310361B (en) * | 2022-08-16 | 2023-09-15 | 中国矿业大学 | Underground coal mine dust concentration prediction method and system based on WGAN-CNN |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111783545A (en) | Coal mine unmanned aerial vehicle image enhancement method based on generation of countermeasure network | |
| CN108108746B (en) | License plate character recognition method based on Caffe deep learning framework | |
| CN107529650B (en) | Closed loop detection method and device and computer equipment | |
| CN110348475B (en) | Confrontation sample enhancement method and model based on spatial transformation | |
| CN109509156B (en) | Image defogging processing method based on generation countermeasure model | |
| CN111126412B (en) | Image key point detection method based on characteristic pyramid network | |
| CN111783551B (en) | Countermeasure sample defense method based on Bayesian convolutional neural network | |
| CN110222604B (en) | Target identification method and device based on shared convolutional neural network | |
| CN112668522B (en) | Human body key point and human body mask joint detection network and method | |
| CN110136162B (en) | Unmanned aerial vehicle visual angle remote sensing target tracking method and device | |
| CN110097522B (en) | Single outdoor image defogging method based on multi-scale convolution neural network | |
| CN108734677B (en) | Blind deblurring method and system based on deep learning | |
| CN112766062B (en) | Human behavior identification method based on double-current deep neural network | |
| CN112766315A (en) | Method and system for testing robustness of artificial intelligence model | |
| Phan et al. | Adversarial imaging pipelines | |
| CN116524183A (en) | Camouflage target detection method based on multitask adapter fine adjustment | |
| CN107274425B (en) | A kind of color image segmentation method and device based on Pulse Coupled Neural Network | |
| CN114021704B (en) | AI neural network model training method and related device | |
| CN112215766B (en) | Image defogging method combining image restoration and image enhancement and convolution network thereof | |
| CN110136164B (en) | Method for removing dynamic background based on online transmission transformation and low-rank sparse matrix decomposition | |
| CN116612355A (en) | Training method and device for face fake recognition model, face recognition method and device | |
| CN114998801A (en) | Forest fire smoke video detection method based on contrast self-supervision learning network | |
| CN108182685B (en) | Image segmentation method and device combining belief propagation idea and PCNN | |
| CN114821629A (en) | Pedestrian re-identification method for performing cross image feature fusion based on neural network parallel training architecture | |
| CN117152687B (en) | Communication line state monitoring system |
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 |