CN112560579A - Obstacle detection method based on artificial intelligence - Google Patents
Obstacle detection method based on artificial intelligence Download PDFInfo
- Publication number
- CN112560579A CN112560579A CN202011307128.4A CN202011307128A CN112560579A CN 112560579 A CN112560579 A CN 112560579A CN 202011307128 A CN202011307128 A CN 202011307128A CN 112560579 A CN112560579 A CN 112560579A
- Authority
- CN
- China
- Prior art keywords
- data set
- image
- generator
- discriminator
- style data
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/35—Categorising the entire scene, e.g. birthday party or wedding scene
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an obstacle detection method based on artificial intelligence. The method comprises the following steps: dividing the collected life scene data set into a second style data set of standard illuminance and a first style data set of other illuminance according to the illuminance; training a loop countermeasure network by taking a set loss function as a target, wherein the loop countermeasure network comprises a first generator, a second generator, a first discriminator and a second discriminator, the first generator is used for converting a first style data set and transmitting the first style data set to the first discriminator, the second generator is used for converting a second style data set and transmitting the second style data set to the second discriminator, and the first discriminator and the second discriminator are used for judging whether a received image accords with a corresponding target style; and processing the image to be detected by using the trained first generator, and outputting the image to a target detection network for obstacle detection. The invention solves the problems of poor robustness and low accuracy of target detection under the complex illumination condition faced by the blind guiding technology.
Description
Technical Field
The invention relates to the technical field of natural image processing, in particular to an obstacle detection method based on artificial intelligence.
Background
According to the statistics of Chinese couplets, at least 500 million blind people exist in China at present, and the number of blind people is increased year by year with the aging of population. By 2050, the number of blind people in the world is expected to reach 7.03 hundred million. The 'guiding blind for the blind' is always a hot research problem, from the traditional guiding stick and the guiding dog to various guiding robots in the present. The traditional blind guiding equipment mainly uses modules such as ultrasonic waves, infrared rays and the like as main sensors, realizes simple obstacle avoidance reminding by detecting the distance and the direction of an obstacle, but has limited information which can be transmitted to the blind; the machine vision blind guiding scheme based on the traditional algorithm is difficult to process the condition that image information of a blind living scene is huge and complex. At present, with the rise of artificial intelligence, the application of machine vision in blind guiding equipment returns to the field of view of the public again, and the appearance of deep learning and convolutional neural networks solves the problem that the traditional machine vision faces the problem that the image is complex and difficult to process to a certain extent.
In recent years, a target detection means based on deep learning has achieved a very good effect in the field of natural image processing, and is widely applied to various fields, such as unmanned application scenes and the like, and the method has a great application value and potential as a blind guiding field for improving the life quality of blind people.
In the prior art, most of blind guiding technologies based on deep target detection upload collected images to a server, then train a network for processing by using a supervised or unsupervised method, and then guide blind by combining other sensing information. The method fully utilizes the advantages of processing complex images by deep learning, and has good performance under the general blind guiding situation. Experiments show that the blind guiding device can accurately identify common objects such as a garbage can, a chair, people and the like in the life scene of the blind through deep learning. However, the existing methods are usually trained under a perfect data set, and the aimed image needs a data set with moderate brightness and reasonable clarity or specially applied in the dark, but has no good robustness to various illumination changes which may occur in the living scene of the blind.
In summary, the drawbacks of the prior art are mainly reflected in: the traditional visual-free method usually applies ultrasonic and infrared sensors, the judgment of the barrier is only limited to the azimuth distance, and the precision is low; the traditional machine vision mainly utilizes a pre-written algorithm to identify the characteristics of a target in an image, and the method has weak migration capability and no intelligence; the machine vision method based on deep learning trains the characteristics of the learning images through the data set, can identify the images of various scenes and carry out target detection, has good detection effect, but the brightness change of the image scenes can also influence the accuracy of the images and increase the training difficulty.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide an obstacle detection method based on artificial intelligence, wherein a novel obstacle detection scheme with robustness under complex illumination is designed by introducing a cyclic generation countermeasure network to preprocess an image.
The invention provides an obstacle detection method based on artificial intelligence, which comprises the following steps:
dividing the collected life scene data set into a second style data set of standard illuminance and a first style data set of other illuminance according to the illuminance;
training a loop countermeasure network by taking a set loss function as a target, wherein the loop countermeasure network comprises a first generator, a second generator, a first discriminator and a second discriminator, the first generator is used for converting a first style data set and transmitting the first style data set to the first discriminator, the second generator is used for converting a second style data set and transmitting the second style data set to the second discriminator, and the first discriminator and the second discriminator are used for judging whether a received image accords with a corresponding target style;
and processing the image to be detected by using the trained first generator, and outputting the image to a target detection network for obstacle detection.
Compared with the prior art, the invention has the advantages that aiming at the characteristic that the illumination intensity of the life scene image of the visually impaired people is complex and changeable, the multi-illumination target detection technology based on artificial intelligence is designed, and the image is preprocessed through the cyclic countermeasure network, so that the problems of poor robustness and low accuracy of target detection under the complex illumination scene faced by the blind guiding technology are solved, and the accurate obstacle detection under the multi-illumination scene is realized.
Other features of the present invention and advantages thereof will become apparent from the following detailed description of exemplary embodiments thereof, which proceeds with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description, serve to explain the principles of the invention.
FIG. 1 is a system framework diagram of an artificial intelligence based obstacle detection method according to one embodiment of the present invention;
fig. 2 is a flowchart of an artificial intelligence based obstacle detection method according to an embodiment of the present invention.
Detailed Description
Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that: the relative arrangement of the components and steps, the numerical expressions and numerical values set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.
The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.
Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.
In all examples shown and discussed herein, any particular value should be construed as merely illustrative, and not limiting. Thus, other examples of the exemplary embodiments may have different values.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, further discussion thereof is not required in subsequent figures.
Referring to fig. 1, the general system framework for obstacle detection based on artificial intelligence provided by the invention comprises an image acquisition module, an image preprocessing module and an object detection module.
The image acquisition module is used for acquiring images of real life scenes, for example, the scene detected by the blind guiding barrier is a common life scene of visually impaired people, the real scene images can be acquired through the camera module, and the images to be detected under different illumination intensities are obtained. In practical application, the camera module can be made into blind guiding glasses, and the user wears the glasses to simulate eyes to see objects in front of the eyes, so that image acquisition is realized.
The image preprocessing module is used for processing images with different illumination intensities, and realizing the illumination intensity normalization of the images so as to improve the subsequent target detection accuracy and improve the robustness aiming at the images with different illumination intensities.
The image preprocessing module is implemented based on a cyclic countermeasure network (CycleGAN) and comprises two generators, denoted GAAnd GBTwo discriminators, denoted DA、DBIn the training process, the loop countermeasure network utilizes two data sets of different styles to perform countermeasure training so as to be applied to image processing of different illumination intensities.
Specifically, blind person life scene data sets containing various illumination intensities are classified, partial self-defined images with standard illumination intensities are selected and classified into a B-style data set, and other images with various illumination intensities are classified into an A-style data set.
Generator GA、GBFor converting an input data set into a different style. For example, the generator is constituted by an encoder, a converter, and a decoder. For generator GAThe encoder extracts features from an input image using a convolutional neural network with the style a dataset as input, and the converter converts feature vectors of the image in the style a dataset into feature vectors in the style B dataset by combining dissimilar features of the image. The decoder uses the deconvolution layer to complete the work of restoring low-level features from the feature vector, and finally the result is obtainedAn image is generated. Similarly, generator GBThe B-style dataset is taken as input to convert the feature vectors in the B-style dataset into feature vectors in the a-style dataset.
Discriminator DA、DBFor determining whether the input image from the respective generator conforms to the target style. The discriminator takes an image as input and tries to predict whether it is the original image or the output image of the generator. The discriminator itself belongs to a convolutional network, and it is necessary to extract features from the image and then determine whether the extracted features belong to a particular class by adding a convolutional layer that produces a one-dimensional output. For example, discriminator DAJudgment GAWhether the generated image inputs B style, if not, the image is discarded, and if so, the image can be added to the B style data set for further training. Similarly, discriminator DBFor judging GBWhether the generated image is input in a style.
Target detection module for receiving via generator GAThe stylized images are processed to enable obstacle detection, such as detecting the type and location of obstacles. The object detection module may employ the Faster RCNN network or use other object detection networks instead.
Preferably, the system framework of fig. 1 further comprises an image normalization process, which aims to perform a preliminary cropping transformation on the captured original image to convert the original image into an image suitable for the loop-through network processing.
In conclusion, the invention utilizes the cyclic countermeasure network, sets generators and discriminators aiming at different illumination intensities, and realizes image style migration through countermeasure training. By the method, the image under any illumination can be converted into a customized standard style, and the image is very friendly to a subsequent target detection network; moreover, two data sets corresponding to each other are not needed for training the loop countermeasure network, and the workload of processing the data sets is remarkably reduced.
To further understand the process of the present invention, fig. 2 shows a flowchart of the obstacle detection method based on artificial intelligence provided by the present invention, which specifically includes the following steps:
and step S210, dividing the collected living scene data set into a data set B of standard illumination and a data set A of other illumination according to the illumination.
The division standard for the data set can be determined according to actual needs, for example, more images with different illumination intensities and different angles are acquired, or a finer-grained division standard is adopted to enhance the accuracy of subsequent detection.
Step S220, training a cyclic countermeasure network by utilizing the data set A and the data set B to obtain a generator G capable of converting images with different illumination intensities into standard illumination intensitiesA。
For example, the cyclic countermeasure network is trained with the set loss function as the optimization target, wherein the loss function setting of the generator and the discriminator can adopt the prior art, and will not be described herein. When training is complete, generator GAThe real-time collected images with different illumination intensities can be converted into standard illumination intensity images.
Step S230, collecting blind living scene images in real time, carrying out standardized processing and transmitting the images to a trained generator GA。
For example, in practical application, the real life scene in front of the eye is collected by using the blind guiding glasses worn, and further through standardization processing, the suitable generator G is obtainedAThe processed image format.
Step S240, the generator GAThe generated image is transmitted to a target detection network for obstacle detection.
Trained generator GAInput images with various illumination intensities and angles can be converted into standard illumination intensity images through style migration, so that the target detection network can accurately detect the types, positions and the like of obstacles.
In conclusion, the obstacle target detection method based on artificial intelligence blind guiding provided by the invention solves the problem of poor effect caused by different illumination influences on obstacle detection in the blind guiding technology, and has strong robustness to illumination change in a day and instantaneous brightness change caused by lamplight which may occur in life scenes of blind people. The target detection is carried out on the life scene images of the blind under different illumination intensities during blind guiding based on the illumination intensity standardization technology of the cyclic countermeasure network, so that data are provided for blind guiding.
The present invention may be a system, method and/or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions embodied therewith for causing a processor to implement various aspects of the present invention.
The computer readable storage medium may be a tangible device that can hold and store the instructions for use by the instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic memory device, a magnetic memory device, an optical memory device, an electromagnetic memory device, a semiconductor memory device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a Static Random Access Memory (SRAM), a portable compact disc read-only memory (CD-ROM), a Digital Versatile Disc (DVD), a memory stick, a floppy disk, a mechanical coding device, such as punch cards or in-groove projection structures having instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media as used herein is not to be construed as transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission medium (e.g., optical pulses through a fiber optic cable), or electrical signals transmitted through electrical wires.
The computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to a respective computing/processing device, or to an external computer or external storage device via a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. The network adapter card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in the respective computing/processing device.
The computer program instructions for carrying out operations of the present invention may be assembler instructions, Instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, aspects of the present invention are implemented by personalizing an electronic circuit, such as a programmable logic circuit, a Field Programmable Gate Array (FPGA), or a Programmable Logic Array (PLA), with state information of computer-readable program instructions, which can execute the computer-readable program instructions.
Aspects of the present invention are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions.
These computer-readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer-readable program instructions may also be stored in a computer-readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer-readable medium storing the instructions comprises an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.
The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions. It is well known to those skilled in the art that implementation by hardware, by software, and by a combination of software and hardware are equivalent.
Having described embodiments of the present invention, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen in order to best explain the principles of the embodiments, the practical application, or improvements made to the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein. The scope of the invention is defined by the appended claims.
Claims (9)
1. An obstacle detection method based on artificial intelligence comprises the following steps:
dividing the collected life scene data set into a second style data set of standard illuminance and a first style data set of other illuminance according to the illuminance;
training a loop countermeasure network by taking a set loss function as a target, wherein the loop countermeasure network comprises a first generator, a second generator, a first discriminator and a second discriminator, the first generator is used for converting a first style data set and transmitting the first style data set to the first discriminator, the second generator is used for converting a second style data set and transmitting the second style data set to the second discriminator, and the first discriminator and the second discriminator are used for judging whether a received image accords with a corresponding target style;
and processing the image to be detected by using the trained first generator, and outputting the image to a target detection network for obstacle detection.
2. The method of claim 1, wherein the first style data set and the second style data set are determined according to the following steps:
acquiring a real scene image through a camera module to obtain images under different illumination intensities;
the life scene data sets containing various illumination intensities are classified, partial self-defined images with standard illumination intensities are selected and classified into a second style data set, and other images with various illumination intensities are classified into a first style data set.
3. The method of claim 1, wherein the first generator and the second generator comprise an encoder, a converter, and a decoder, the encoder extracting features from the input image using a convolutional neural network; the converter is used for converting the characteristic vectors in the input style data set by combining different characteristics of the images; the decoder uses the deconvolution layer to restore low-level features from the feature vectors to obtain a generated image.
4. The method of claim 1, wherein the first discriminator and the second discriminator employ a convolutional network to extract features from the image and determine whether the extracted features belong to a target style class by adding convolutional layers that produce a one-dimensional output.
5. The method of claim 1, wherein the first style data set and the second style data set are image data suitable for processing by the loop countermeasure network after performing cropping transformation on the captured life scene original.
6. The method of claim 1, wherein processing the image to be detected with the trained first generator and outputting to the target detection network for obstacle detection comprises:
acquiring image data of a life scene of the blind as an image to be detected by using blind guiding glasses, wherein the blind guiding glasses comprise a camera module;
the image to be detected is subjected to standardization processing and then is transmitted to a trained first generator;
the first generator transmits the output result to the target detection network to identify the obstacle type and position.
7. The method of claim 1, wherein the object detection network is a fast RCNN network.
8. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.
9. A computer device comprising a memory and a processor, on which memory a computer program is stored which is executable on the processor, characterized in that the steps of the method of any of claims 1 to 7 are implemented when the processor executes the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011307128.4A CN112560579A (en) | 2020-11-20 | 2020-11-20 | Obstacle detection method based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011307128.4A CN112560579A (en) | 2020-11-20 | 2020-11-20 | Obstacle detection method based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112560579A true CN112560579A (en) | 2021-03-26 |
Family
ID=75044112
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011307128.4A Pending CN112560579A (en) | 2020-11-20 | 2020-11-20 | Obstacle detection method based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112560579A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113065478A (en) * | 2021-04-09 | 2021-07-02 | 长安大学 | Complex weather road vehicle target detection method |
| CN113554087A (en) * | 2021-07-20 | 2021-10-26 | 清华大学深圳国际研究生院 | Archaeological graph automatic generation method based on generation type confrontation network |
| CN113688825A (en) * | 2021-05-17 | 2021-11-23 | 海南师范大学 | AI intelligent garbage recognition and classification system and method |
| CN114065838A (en) * | 2021-10-22 | 2022-02-18 | 中国科学院深圳先进技术研究院 | Low-illumination obstacle detection method, system, terminal and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190147343A1 (en) * | 2017-11-15 | 2019-05-16 | International Business Machines Corporation | Unsupervised anomaly detection using generative adversarial networks |
| CN111899251A (en) * | 2020-08-06 | 2020-11-06 | 中国科学院深圳先进技术研究院 | Copy-move type forged image detection method for distinguishing forged source and target area |
-
2020
- 2020-11-20 CN CN202011307128.4A patent/CN112560579A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190147343A1 (en) * | 2017-11-15 | 2019-05-16 | International Business Machines Corporation | Unsupervised anomaly detection using generative adversarial networks |
| CN111899251A (en) * | 2020-08-06 | 2020-11-06 | 中国科学院深圳先进技术研究院 | Copy-move type forged image detection method for distinguishing forged source and target area |
Non-Patent Citations (2)
| Title |
|---|
| 余广民 等: "基于GAN网络的异常检测算法研究" * |
| 赵亮 等: "基于循环生成对抗网络的光照补偿方法" * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113065478A (en) * | 2021-04-09 | 2021-07-02 | 长安大学 | Complex weather road vehicle target detection method |
| CN113688825A (en) * | 2021-05-17 | 2021-11-23 | 海南师范大学 | AI intelligent garbage recognition and classification system and method |
| CN113554087A (en) * | 2021-07-20 | 2021-10-26 | 清华大学深圳国际研究生院 | Archaeological graph automatic generation method based on generation type confrontation network |
| CN113554087B (en) * | 2021-07-20 | 2023-11-10 | 清华大学深圳国际研究生院 | Automatic archaeological line graph generation method based on generation type countermeasure network |
| CN114065838A (en) * | 2021-10-22 | 2022-02-18 | 中国科学院深圳先进技术研究院 | Low-illumination obstacle detection method, system, terminal and storage medium |
| WO2023065497A1 (en) * | 2021-10-22 | 2023-04-27 | 中国科学院深圳先进技术研究院 | Low-illumination obstacle detection method and system, and terminal and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Adaptive fusion for RGB-D salient object detection | |
| CN112560579A (en) | Obstacle detection method based on artificial intelligence | |
| CN108229504B (en) | Image analysis method and device | |
| US11393256B2 (en) | Method and device for liveness detection, and storage medium | |
| Liu et al. | A cross-modal adaptive gated fusion generative adversarial network for RGB-D salient object detection | |
| WO2021031954A1 (en) | Object quantity determination method and apparatus, and storage medium and electronic device | |
| CN111783646B (en) | Training method, device, equipment and storage medium of pedestrian re-identification model | |
| CN112861575A (en) | Pedestrian structuring method, device, equipment and storage medium | |
| US10169661B2 (en) | Filtering methods for visual object detection | |
| CN113591770A (en) | Multimode fusion obstacle detection method and device based on artificial intelligence blind guiding | |
| US20220148292A1 (en) | Method for glass detection in real scenes | |
| CN111626956A (en) | Image deblurring method and device | |
| WO2019214321A1 (en) | Vehicle damage identification processing method, processing device, client and server | |
| Shah et al. | Efficient portable camera based text to speech converter for blind person | |
| KR20220153667A (en) | Feature extraction methods, devices, electronic devices, storage media and computer programs | |
| CN112270745A (en) | Image generation method, device, equipment and storage medium | |
| CN112529930A (en) | Context learning medical image segmentation method based on focus fusion | |
| CN112949451B (en) | Cross-modal target tracking method and system through modal perception feature learning | |
| CN113570689A (en) | Portrait cartoon method, apparatus, medium and computing device | |
| CN116580353A (en) | Automatic early warning method and system for potential safety hazards of thermal power plant | |
| CN114038067B (en) | Coal mine personnel behavior detection method, equipment and storage medium | |
| CN114612971A (en) | Face detection method, model training method, electronic device, and program product | |
| CN114550291A (en) | Gait feature extraction method, device and equipment | |
| CN114842411A (en) | Group behavior identification method based on complementary space-time information modeling | |
| CN114820885A (en) | Image editing method and model training method, device, equipment and medium thereof |
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 |