CN111464738A - Image command device for question and answer based on deep neural network and attention mechanism - Google Patents
Image command device for question and answer based on deep neural network and attention mechanism Download PDFInfo
- Publication number
- CN111464738A CN111464738A CN202010247842.2A CN202010247842A CN111464738A CN 111464738 A CN111464738 A CN 111464738A CN 202010247842 A CN202010247842 A CN 202010247842A CN 111464738 A CN111464738 A CN 111464738A
- Authority
- CN
- China
- Prior art keywords
- neural network
- image
- module
- attention mechanism
- electrically connected
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/61—Control of cameras or camera modules based on recognised objects
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield 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
- G06N3/045—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
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
Abstract
The invention discloses a commanding device for image question answering based on a deep neural network and an attention mechanism, which belongs to the technical field of image question answering and comprises a base, a display screen, an electric telescopic rod, a control box and universal wheels, wherein a storage battery is fixedly arranged in the middle of the top of the base, hydraulic lifting devices are fixedly arranged at four corners of the top of the base, a workbench is fixedly arranged at the top of the hydraulic lifting devices, the display screen is fixedly arranged at the left side of the top of the workbench, the structure of the commanding device is reasonable, the angle and the height of a high-definition camera are conveniently adjusted through the matching of the electric telescopic rod, the mounting seat, the mounting frame, the high-definition camera and an angle adjusting knob, the integrity of image collection is ensured, and through the matching of a convolutional neural network module, a central processing unit, an attribute detector, a recursive neural network module, an automatic image, the accuracy of computer image question answering is improved.
Description
Technical Field
The invention relates to the technical field of image question answering, in particular to a commanding device for image question answering based on a deep neural network and an attention mechanism.
Background
With the current popularization and development of the internet, multimedia data (including images, videos and the like) spread on the basis of the internet also shows an exponential growth trend. In the face of such huge image information resources, how to effectively and reasonably manage the information by using a computer is a research with practical significance and urgent need. Moreover, computer vision has become more and more popular in society, and has been widely applied to the fields of image retrieval, intelligent monitoring of unmanned aerial vehicles, unmanned vehicles and the like, and the core technology of the applications is image understanding. Image understanding is the study of how to make a computer understand the contents in an image, implementing a computer system like the human visual system, for understanding the outside world. According to the latest leading-edge research, image understanding can be divided into four levels:
1. image classification: the picture only contains a single object, and the computer needs to identify the type of the object;
2. target detection: the picture comprises a plurality of objects, the proportion of the area of each object in the picture is not fixed, and a computer not only needs to detect the position of each object, but also needs to identify the type of each object;
3. image description: the computer generates a sentence describing the content in the picture;
4. image question answering: the input has a question related to the image (the answer can be obtained according to the content of the picture) besides the image, and the computer needs to answer the question according to the content of the picture.
Disclosure of Invention
This section is for the purpose of summarizing some aspects of embodiments of the invention and to briefly introduce some preferred embodiments. In this section, as well as in the abstract and the title of the invention of this application, simplifications or omissions may be made to avoid obscuring the purpose of the section, the abstract and the title, and such simplifications or omissions are not intended to limit the scope of the invention.
The invention is provided in view of the problems existing in the prior image question-answering commanding device.
Therefore, the invention aims to provide the image command device for questioning and answering based on the deep neural network and the attention mechanism, which can realize that the camera can conveniently acquire images of all angles in the using process, and the image command device has higher precision.
To solve the above technical problem, according to an aspect of the present invention, the present invention provides the following technical solutions:
the image question and answer commanding device based on the deep neural network and the attention mechanism comprises a base, a display screen, an electric telescopic rod, a control box and universal wheels, wherein a storage battery is fixedly mounted in the middle of the top of the base, a hydraulic lifting device is fixedly mounted at four corners of the top of the base, a workbench is fixedly mounted at the top of the hydraulic lifting device, the display screen is fixedly mounted at the left side of the top of the workbench, the electric telescopic rod is fixedly mounted at the right side of the top of the workbench, a mounting seat is fixedly mounted at the top of the electric telescopic rod, a servo motor is fixedly mounted in the middle of the bottom of the mounting seat, a rotary disc is arranged at the top output end of the servo motor through the mounting seat, mounting frames are fixedly mounted at two sides of the top of the rotary disc, a high-definition camera is rotatably connected onto the mounting frames, the inner cavity of the control box is respectively and fixedly provided with a central processing unit, a convolutional neural network module, an attribute detector, a recurrent neural network module, an automatic image description module, an attribute prediction module, a signal receiving module and a driver, wherein the central processing unit is electrically connected with the convolutional neural network module in an input mode, the convolutional neural network module is electrically connected with a high-definition camera in an input mode, the central processing unit is electrically connected with the recurrent neural network module in an output mode, the recurrent neural network module is electrically connected with the attribute prediction module in an output mode, the attribute prediction module is electrically connected with the driver in an output mode, the central processing unit is connected with the attribute detector in a two-way mode, the recurrent neural network module is connected with the automatic image description module in a two-way mode, the central processing unit is electrically, the electric output of the central processing unit is connected with the driver, and the electric output of the driver is connected with the electric telescopic rod and the servo motor.
As a preferable embodiment of the image command device for question and answer based on the deep neural network and the attention mechanism according to the present invention, wherein: the bottom four corners fixed mounting of base has the universal wheel, fixed mounting has the stopper on the universal wheel.
As a preferable embodiment of the image command device for question and answer based on the deep neural network and the attention mechanism according to the present invention, wherein: the handheld terminal is a remote controller or a smart phone.
As a preferable embodiment of the image command device for question and answer based on the deep neural network and the attention mechanism according to the present invention, wherein: a storage drawer is fixedly arranged in the middle of the bottom of the workbench.
As a preferable embodiment of the image command device for question and answer based on the deep neural network and the attention mechanism according to the present invention, wherein: and a heat dissipation seat is arranged between the display screen and the workbench.
As a preferable embodiment of the image command device for question and answer based on the deep neural network and the attention mechanism according to the present invention, wherein: the mounting bracket with the junction of high definition digtal camera is provided with angle adjust knob.
As a preferable embodiment of the image command device for question and answer based on the deep neural network and the attention mechanism according to the present invention, wherein: the attribute detector detects an attribute of the image using a predictor.
As a preferable embodiment of the image command device for question and answer based on the deep neural network and the attention mechanism according to the present invention, wherein: the automatic image description module adopts a sentence generation model based on a long-time memory network.
As a preferable embodiment of the image command device for question and answer based on the deep neural network and the attention mechanism according to the present invention, wherein: the storage battery is connected with the control box and the display screen through electric wires.
Compared with the prior art, the invention has the beneficial effects that: through this image command device for questioning and answering based on degree of depth neural network and attention mechanism's setting, structural design is reasonable, through electric telescopic handle, the mount pad, the mounting bracket, high definition digtal camera and angle adjust knob's cooperation, conveniently adjust high definition digtal camera's angle and height, guarantee image acquisition's integrality, through the cooperation of convolution neural network module, central processing unit, attribute detector, recursion neural network module, automatic image description module and attribute prediction module, the accurate nature of computer image questioning and answering has been improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the present invention will be described in detail with reference to the accompanying drawings and detailed embodiments, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise. Wherein:
FIG. 1 is a schematic structural view of the present invention;
FIG. 2 is a schematic top view of the present invention;
FIG. 3 is a block diagram of the system of the present invention.
In the figure; 100, 110 storage batteries, 120 hydraulic lifting devices, 130, 140 storage drawers, 200 display screens, 210 heat sinks, 300 electric telescopic rods, 310 mounting seats, 320 servo motors, 330 turntables, 340 mounting frames, 350 high-definition cameras, 360-degree adjusting knobs, 400 control boxes, 410 central processing units, 420 convolutional neural network modules, 430 attribute detectors, 440 recurrent neural network modules, 450 automatic image description modules, 460 attribute prediction modules, 470 signal receiving modules, 480 drivers, 490 handheld terminals, 500 universal wheels and 510 limiters.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways than those specifically described herein, and it will be apparent to those of ordinary skill in the art that the present invention may be practiced without departing from the spirit and scope of the present invention, and therefore the present invention is not limited to the specific embodiments disclosed below.
Next, the present invention will be described in detail with reference to the drawings, wherein for convenience of illustration, the cross-sectional view of the device structure is not enlarged partially according to the general scale, and the drawings are only examples, which should not limit the scope of the present invention. In addition, the three-dimensional dimensions of length, width and depth should be included in the actual fabrication.
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
The invention provides the following technical scheme: the image command device for question and answer based on the deep neural network and the attention mechanism can realize that a camera can conveniently acquire images of all angles in the using process, the image command device is higher in precision of question and answer, and please refer to fig. 1 to 3, and the image command device comprises a base 100, a display screen 200, an electric telescopic rod 300, a control box 400 and a universal wheel 500;
referring to fig. 1 again, a storage battery 110 is fixedly installed in the middle of the top of the base 100, hydraulic lifting devices 120 are fixedly installed at four corners of the top of the base 100, a workbench 130 is fixedly installed at the top of the hydraulic lifting devices 120, a storage drawer 140 is fixedly installed in the middle of the bottom of the workbench 130, specifically, the storage battery 110 is adhered in the middle of the top of the base 100, the hydraulic lifting devices 120 are welded at four corners of the top of the base 100, the workbench 130 is welded at the top of the hydraulic lifting devices 120, the storage drawer 140 is slidably connected in the middle of the bottom of the workbench 130, the base 100 is used for bearing the storage battery, the electric power tool comprises a hydraulic lifting device 120 and universal wheels 500, a storage battery 110 is used for improving electric energy and ensuring power supply, the hydraulic lifting device 120 is used for bearing a workbench 130 and adjusting the height, the workbench 130 is used for bearing a display screen 200 and an electric telescopic rod 300, and a storage drawer 140 is used for placing tools;
referring to fig. 1 again, a display screen 200 is fixedly mounted on the left side of the top of the workbench 130, a heat dissipation seat 210 is disposed between the display screen 200 and the workbench 130, specifically, the display screen 200 is used for displaying the image question and answer result, and the heat dissipation seat 210 is used for improving heat dissipation performance and safety;
referring to fig. 1 again, an electric telescopic rod 300 is fixedly installed on the right side of the top of the workbench 130, an installation seat 310 is fixedly installed on the top of the electric telescopic rod 300, a servo motor 320 is fixedly installed in the middle of the bottom of the installation seat 310, a rotary table 330 is arranged at the top output end of the servo motor 320 in a penetrating mode through the installation seat 310, installation seats 340 are fixedly installed on two sides of the top of the rotary table 330, a high-definition camera 350 is rotatably connected to the installation seats 340 through a rotary shaft, specifically, the electric telescopic rod 300 is welded on the right side of the top of the workbench 130, the installation seat 310 is welded on the top of the electric telescopic rod 300, the servo motor 320 is screwed in the middle of the bottom of the installation seat 310, the top output end of the servo motor 320 is rotatably connected to the rotary table 330 in a penetrating mode through the installation seat 310, the installation seats 340 are welded on two sides of, the electric telescopic rod 300 is used for bearing the mounting seat 310 and adjusting the height, the mounting seat 310 is used for bearing the servo motor 320, the servo motor 320 is used for driving the turntable 330 to rotate, so that the image acquisition angle is adjusted, the mounting frame 340 is used for bearing the high-definition camera 350, the high-definition camera 350 is used for acquiring images, and the angle adjusting knob 360 is used for adjusting the angle of the high-definition camera 350;
referring to fig. 1 and fig. 3 again, the top of the display screen 200 is fixedly installed with a control box 400, the inner cavity of the control box 400 is respectively fixedly installed with a central processor 410, a convolutional neural network module 420, an attribute detector 430, a recurrent neural network module 440, an automatic image description module 450, an attribute prediction module 460, a signal receiving module 470 and a driver 480, the central processor 410 is electrically connected with the convolutional neural network module 420 in an input manner, the convolutional neural network module 420 is electrically connected with the high-definition camera 350 in an input manner, the central processor 410 is electrically connected with the recurrent neural network module 440 in an output manner, the recurrent neural network module 440 is electrically connected with the attribute prediction module 460 in an output manner, the attribute prediction module 460 is electrically connected with the driver 480 in an output manner, the central processor 410 is connected with the attribute detector 430 in a bidirectional manner, the recurrent neural network module 440 is connected with the automatic image, the signal receiving module 470 is electrically connected with the handheld terminal 490 in an input manner, the central processing unit 410 is electrically connected with the driver 480 in an output manner, the driver 480 is electrically connected with the electric telescopic rod 300 and the servo motor 320 in an output manner, specifically, the top of the display screen 200 is adhered with the control box 400, the inner cavity of the control box 400 is respectively adhered with the central processing unit 410, the convolutional neural network module 420, the attribute detector 430, the recurrent neural network module 440, the automatic image description module 450, the attribute prediction module 460, the signal receiving module 470 and the driver 480 in an inner cavity, the central processing unit 410 is electrically connected with the convolutional neural network module 420 in an input manner, the convolutional neural network module 420 is electrically connected with the high-definition camera 350 in an input manner, the central processing unit 410 is electrically connected with the recurrent neural network module 440 in an output manner, the recurrent neural network module 440, the central processing unit 410 is connected with the attribute detector 430 in a bidirectional manner, the recurrent neural network module 440 is connected with the automatic image description module 450 in a bidirectional manner, the central processing unit 410 is electrically connected with the signal receiving module 470 in an input manner, the signal receiving module 470 is electrically connected with the handheld terminal 490 in an input manner, the central processing unit 410 is electrically connected with the driver 480 in an output manner, the driver 480 is electrically connected with the electric telescopic rod 300 and the servo motor 320 in an output manner, the control box 400 is used for bearing electronic components such as the central processing unit 410, the central processing unit 410 is used for receiving circuit signals received by the signal receiving module 470 and the convolutional neural network module 420, the convolutional neural network module 420 is used for receiving and collecting image characteristics in images collected by the high-definition camera 350, the attribute detector 430 is used for collecting key characteristics in the image characteristics, the recurrent neural network, the automatic image description module is used for describing the acquired image characteristics for multiple times, the attribute prediction module 460 is used for calculating the most reasonable answer by combining the image characteristics and the problem characteristics, the handheld terminal 490 is a smart phone and is used for sending signals and inputting problems, the signal receiving module 470 is used for receiving circuit signals sent by the handheld terminal 490, and the driver 480 is used for controlling the electric telescopic rod 300 and the servo motor 320 to work, so that the acquisition angle height of the high-definition camera 350 can be conveniently adjusted;
referring to fig. 1 again, the universal wheels 500 are fixedly installed at four corners of the bottom of the base 100, the stoppers 510 are fixedly installed on the universal wheels 500, specifically, the universal wheels 500 are screwed at four corners of the bottom of the base 100, the stoppers 510 are screwed on the universal wheels 500, the universal wheels 500 are used for facilitating the movement of the device, and the stoppers 510 are used for fixing the universal wheels 500.
The working principle is as follows: in the process of using the image commanding device for question and answer based on the deep neural network and the attention mechanism, firstly, the high-definition camera 350 is adjusted to a proper image acquisition angle through the cooperation of the electric telescopic rod 300, the mounting seat 310, the servo motor 320, the mounting frame 340 and the angle adjusting knob 360, when the high-definition camera 350 acquires an image, the image is transmitted to the convolutional neural network module 420, the image features in the image are extracted through the convolutional neural network module, then the extracted features are sent to the central processing unit 410 to be repeatedly compared with the attribute detector 430, and the most reasonable answer is pre-calculated through the attribute budget module by combining the image features and the question features after the extracted features are sent to the recursive neural network module 440.
While the invention has been described above with reference to an embodiment, various modifications may be made and equivalents may be substituted for elements thereof without departing from the scope of the invention. In particular, the various features of the disclosed embodiments of the invention may be used in any combination, provided that no structural conflict exists, and the combinations are not exhaustively described in this specification merely for the sake of brevity and resource conservation. Therefore, it is intended that the invention not be limited to the particular embodiments disclosed, but that the invention will include all embodiments falling within the scope of the appended claims.
Claims (9)
1. Image is commander for questioning and answering based on degree of depth neural network and attention mechanism, its characterized in that: the multifunctional electric lifting table comprises a base (100), a display screen (200), an electric telescopic rod (300), a control box (400) and universal wheels (500), wherein a storage battery (110) is fixedly mounted in the middle of the top of the base (100), a hydraulic lifting device (120) is fixedly mounted at the four corners of the top of the base (100), a workbench (130) is fixedly mounted at the top of the hydraulic lifting device (120), the display screen (200) is fixedly mounted at the left side of the top of the workbench (130), the electric telescopic rod (300) is fixedly mounted at the right side of the top of the workbench (130), a mounting seat (310) is fixedly mounted at the top of the electric telescopic rod (300), a servo motor (320) is fixedly mounted in the middle of the bottom of the mounting seat (310), the top output end of the servo motor (320) penetrates through the mounting seat (310) to be provided with a turntable (330), and mounting racks (340, the high-definition camera (350) is rotatably connected to the mounting rack (340) through a rotating shaft, the control box (400) is fixedly mounted at the top of the display screen (200), a central processing unit (410), a convolutional neural network module (420), an attribute detector (430), a recurrent neural network module (440), an automatic image description module (450), an attribute prediction module (460), a signal receiving module (470) and a driver (480) are respectively and fixedly mounted in an inner cavity of the control box (400), the central processing unit (410) is electrically connected with the convolutional neural network module (420) in an input mode, the convolutional neural network module (420) is electrically connected with the high-definition camera (350) in an input mode, the central processing unit (410) is electrically connected with the recurrent neural network module (440) in an output mode, and the recursive neural network module (440) is electrically connected with the attribute prediction module (460, the attribute prediction module (460) is electrically connected with the driver (480), the central processing unit (410) is connected with the attribute detector (430) in a bidirectional mode, the recurrent neural network module (440) is connected with the automatic image description module (450) in a bidirectional mode, the central processing unit (410) is electrically connected with the signal receiving module (470) in an input mode, the signal receiving module (470) is electrically connected with the handheld terminal (490) in an input mode, the central processing unit (410) is electrically connected with the driver (480) in an output mode, and the driver (480) is electrically connected with the electric telescopic rod (300) and the servo motor (320) in an output mode.
2. The image command device for quiz based on deep neural network and attention mechanism according to claim 1, wherein: the bottom four corners fixed mounting of base (100) has universal wheel (500), fixed mounting has stopper (510) on universal wheel (500).
3. The image command device for quiz based on deep neural network and attention mechanism according to claim 1, wherein: the handheld terminal (490) is a remote controller or a smart phone.
4. The image command device for quiz based on deep neural network and attention mechanism according to claim 1, wherein: a storage drawer (140) is fixedly arranged in the middle of the bottom of the workbench (130).
5. The image command device for quiz based on deep neural network and attention mechanism according to claim 1, wherein: a heat dissipation seat (210) is arranged between the display screen (200) and the workbench (130).
6. The image command device for quiz based on deep neural network and attention mechanism according to claim 1, wherein: the joint of the mounting rack (340) and the high-definition camera (350) is provided with an angle adjusting knob (360).
7. The image command device for quiz based on deep neural network and attention mechanism according to claim 1, wherein: the property detector (430) detects properties of the image using a predictor.
8. The image command device for quiz based on deep neural network and attention mechanism according to claim 1, wherein: the automatic image description module (450) employs a sentence generation model based on a long-and-short memory network.
9. The image command device for quiz based on deep neural network and attention mechanism according to claim 1, wherein: the storage battery (110) is connected with the control box (400) and the display screen (200) through wires.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010247842.2A CN111464738A (en) | 2020-04-01 | 2020-04-01 | Image command device for question and answer based on deep neural network and attention mechanism |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010247842.2A CN111464738A (en) | 2020-04-01 | 2020-04-01 | Image command device for question and answer based on deep neural network and attention mechanism |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111464738A true CN111464738A (en) | 2020-07-28 |
Family
ID=71680993
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010247842.2A Pending CN111464738A (en) | 2020-04-01 | 2020-04-01 | Image command device for question and answer based on deep neural network and attention mechanism |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111464738A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112002382A (en) * | 2020-08-26 | 2020-11-27 | 四川大学华西第二医院 | Health care follow-up visit and medical record management system for children |
| CN112270985A (en) * | 2020-11-25 | 2021-01-26 | 上海工程技术大学 | Physique identification system for health management |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107076567A (en) * | 2015-05-21 | 2017-08-18 | 百度（美国）有限责任公司 | Multilingual image question and answer |
| CN207099259U (en) * | 2017-08-31 | 2018-03-13 | 河北科技大学 | A kind of new Visual Communication Design platform |
| CN108170816A (en) * | 2017-12-31 | 2018-06-15 | 厦门大学 | A kind of intelligent vision Question-Answering Model based on deep neural network |
| CN109540793A (en) * | 2018-10-26 | 2019-03-29 | 陈浩 | A kind of vision inspection apparatus with far field vision image transmitting |
-
2020
- 2020-04-01 CN CN202010247842.2A patent/CN111464738A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107076567A (en) * | 2015-05-21 | 2017-08-18 | 百度（美国）有限责任公司 | Multilingual image question and answer |
| CN207099259U (en) * | 2017-08-31 | 2018-03-13 | 河北科技大学 | A kind of new Visual Communication Design platform |
| CN108170816A (en) * | 2017-12-31 | 2018-06-15 | 厦门大学 | A kind of intelligent vision Question-Answering Model based on deep neural network |
| CN109540793A (en) * | 2018-10-26 | 2019-03-29 | 陈浩 | A kind of vision inspection apparatus with far field vision image transmitting |
Non-Patent Citations (1)
| Title |
|---|
| 李庆: "基于深度神经网络和注意力机制的图像问答研究", 《万方平台》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112002382A (en) * | 2020-08-26 | 2020-11-27 | 四川大学华西第二医院 | Health care follow-up visit and medical record management system for children |
| CN112270985A (en) * | 2020-11-25 | 2021-01-26 | 上海工程技术大学 | Physique identification system for health management |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111464738A (en) | Image command device for question and answer based on deep neural network and attention mechanism | |
| CN105323488A (en) | An intelligent camera device | |
| CN211181120U (en) | Education informatization special teaching experiment platform | |
| CN217690323U (en) | Emergency medicine multimedia teaching practical training equipment | |
| CN211604004U (en) | Internet-based intelligent terminal for manufacturing data transmission | |
| CN112001864A (en) | Image training device based on deep learning | |
| CN212482394U (en) | Flexible stereoscopic vision measuring device for target space coordinates | |
| CN214669375U (en) | Electric automation equipment detection device | |
| CN205945963U (en) | On -vehicle panoramic camera system | |
| CN209964214U (en) | Intelligent 3D binocular sensor | |
| CN204463152U (en) | Books in libraries Data Enter device | |
| CN211628458U (en) | Networked teaching platform based on big data mining technology | |
| CN213339126U (en) | Device for vehicle identification and track determination | |
| CN218455151U (en) | Omnibearing photogrammetric device | |
| CN214154614U (en) | Target tracker based on video | |
| CN104700063A (en) | Library book information entering device | |
| CN210627484U (en) | Road evacuation device for intelligent traffic | |
| CN210972469U (en) | Intelligent garbage recycling device with Beidou positioning chip | |
| CN213443127U (en) | Unmanned aerial vehicle aerial photography device based on DOM generation technology | |
| CN213028347U (en) | Network computer lab video monitoring device | |
| CN211787581U (en) | Mathematical education equipment based on computer | |
| CN214044130U (en) | Anti-drop waterproof and dustproof sleeve for USB connection of human-computer equipment | |
| CN214042613U (en) | English interesting teaching board for primary school teaching | |
| CN218336198U (en) | Electric power planning information management system | |
| CN211906514U (en) | Mobile interconnection operation and maintenance monitoring device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200728 |
|
| RJ01 | Rejection of invention patent application after publication |