CN111312022A - Intelligent scoring device and method for chemical carbon dioxide preparation experiment - Google Patents
Intelligent scoring device and method for chemical carbon dioxide preparation experiment Download PDFInfo
- Publication number
- CN111312022A CN111312022A CN202010203120.7A CN202010203120A CN111312022A CN 111312022 A CN111312022 A CN 111312022A CN 202010203120 A CN202010203120 A CN 202010203120A CN 111312022 A CN111312022 A CN 111312022A
- Authority
- CN
- China
- Prior art keywords
- video
- deep learning
- neural network
- network model
- carbon dioxide
- 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
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B19/00—Teaching not covered by other main groups of this subclass
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses an intelligent scoring device and method for a chemical carbon dioxide preparation experiment, which comprises a transparent glass vessel with a color circle label, a black experiment table top for reducing the light reflection degree, a fisheye lens for expanding the video field, a recording scheme of double-lens cooperative work, an efficient and accurate deep learning neural network model and a computer for rigorous and accurate scoring logic judgment. The invention utilizes the deep learning neural network model, and the hardware device provided by the invention is matched, so that the dependence of an examination site on the number of teachers can be reduced, the workload of the teachers is also reduced, the accuracy of the evaluation of the experimental examination is improved, the experimental process of each student is filed, if the examination result is questioned, the teachers and the students can also carry out reexamination through the reserved video, the burden of the teachers and the workers can be reduced at the same time, the evaluation efficiency and the evaluation accuracy can be improved, and the interest of the students in chemical experiments can be improved.
Description
Technical Field
The invention relates to the technical field of teaching and examination of experiments in education and middle schools, in particular to an intelligent scoring device and method for experiments of chemical carbon dioxide preparation.
Background
In the teaching and examination process of the middle school experiment, a teacher needs to guide or judge the experiment operation of students. Especially chemistry, the experiment focusing on the experimental process and watching the practical ability of students is time-consuming and labor-consuming if a teacher supervises the experiment on site one by one, and particularly when a lot of students exist, a teacher needs to guide or judge the experimental operation of the students on site for a long time. Due to the time problem, many students cannot obtain one-to-one guidance, the energy of teachers is also greatly consumed, and meanwhile, the situation that misjudgment cannot be carried out for leaving the evidence exists. Therefore, how to reduce the teaching pressure of teachers and make the experimental operation of each student be effectively guided and accurately judged is a problem that needs to be solved by those skilled in the art.
To date, a variety of deep learning frameworks such as deep neural networks, convolutional neural networks, deep belief networks, and recurrent neural networks have been successfully applied to the fields of computer vision, speech recognition, natural language processing, audio recognition, and bioinformatics, and have achieved excellent results, but in the education industry, there have been few sophisticated techniques developed to combine deep learning with conventional education equipment.
Disclosure of Invention
The invention aims to provide an intelligent scoring device and method for a chemical carbon dioxide preparation experiment, which can not only reduce the burden of teachers and workers, but also improve the judgment efficiency and the judgment accuracy so as to solve the problems in the background technology.
In order to achieve the purpose, the invention provides the following technical scheme:
an intelligent scoring device for a chemical carbon dioxide preparation experiment comprises a transparent glassware, a black experiment table top for reducing the light reflection degree, a fisheye lens for expanding the video visual field, a deep learning neural network model, a computer for rigorous and accurate scoring logic judgment and an equipment bracket matched with the transparent glassware; the equipment support is used for clamping and fixing a transparent glassware to perform carbon dioxide experiment preparation on a black experiment table top, and different color ring labels are arranged on the equipment support and the transparent glassware; the fisheye lens comprises a front-view lens and a top-view lens, the front-view lens is arranged on the black experiment table top in a right-facing mode, and the top-view lens is hung and arranged above the black experiment table top through a hanging bracket; the deep learning neural network model is placed on a black experiment desktop and is connected with a computer, and the computer is placed above the deep learning neural network model.
Furthermore, the transparent glassware comprises a beaker, a glass rod, a test tube, a 90-degree teaching tube and a gas collecting bottle, color ring labels with different colors are embedded on the glass surfaces of different utensils, and the transparent glassware corresponding to the color ring label with each color is recorded into the deep learning neural network model.
Furthermore, the black experiment table top is made of a black light absorption material.
Furthermore, the fisheye lens transmits the recorded pictures into a deep learning neural network model, and a corresponding fisheye correction algorithm is programmed in the deep learning neural network model.
Further, the deep learning neural network model is composed of three parts of a standard SSD network, a lightweight scratch network and a bidirectional network.
The invention provides another technical scheme: an intelligent scoring method for chemical carbon dioxide preparation experiments comprises the following steps:
s1: placing the transparent glassware with the color ring label on a black experiment table top, standing a student in front of the experiment table, and completing corresponding experiment operation according to experiment requirements;
s2: monitoring each operation flow of students by the cooperative work of a front-view lens and a double-camera of a top-view lens in a fisheye lens for expanding the visual field of a video;
s3: the corrected forward-looking video and the normal top-looking video are transmitted to a computer together, the computer firstly cleans the acquired video to obtain a better preprocessing effect and specifies the video into a specification which can be transmitted to a deep learning neural network model, wherein the data cleaning mainly comprises fisheye correction of the forward-looking video;
s4: after the video enters the deep learning neural network model, a video with a detection effect is returned, and meanwhile, two video stream files containing all video information are provided, wherein the two video stream files are processing data of a top view video and a front view video, and the processing data contain the identification detection result of the chemical equipment of each frame in the video;
s5: and (4) performing logic relation judgment scoring according to the position relation of the transparent glassware and the equipment support through the video stream file obtained in the S4 and entering a scoring algorithm of a computer.
Furthermore, in S3, before fisheye correction, the front-view lens is calibrated to obtain the camera internal and external parameters, and then fisheye correction is performed through a fisheye correction algorithm in the computer, so as to obtain a normal video image.
Compared with the prior art, the invention has the beneficial effects that:
according to the intelligent scoring device and method for the chemical carbon dioxide preparation experiment, the deep learning neural network model is used, the transparent glassware with the color ring label, the experiment table top for reducing the light reflection degree and the fisheye lens for expanding the video field of view are matched, the dependence of an examination site on the number of teachers can be reduced, the workload of the teachers is reduced, the accuracy of the experimental examination scoring is improved, the experimental process of each student is filed, if the examination score is questioned, the teachers and the students can perform rechecking after the fact through the reserved video, the burdens of the teachers and the workers can be reduced, the judging efficiency and the judging accuracy can be improved, and the interest of the students in the chemical experiment can be improved.
Drawings
FIG. 1 is a schematic view of the overall structure of the scoring device according to the present invention;
FIG. 2 is a block diagram of an equipment rack of the present invention;
FIG. 3 is a view of the structure of a sealing plug in the transparent glassware article of the present invention;
FIG. 4 is a view showing the structure of a cuvette in a transparent glass vessel according to the present invention;
FIG. 5 is a diagram of a beaker of clear glassware according to the present invention;
FIG. 6 is a flowchart of the scoring method of the present invention.
In the figure: 1. a transparent glassware; 101. a color circle label; 2. a black experimental table top; 3. a fisheye lens; 301. a forward-looking lens; 302. a top view lens; 4. deep learning neural network models; 5. a computer; 6. an equipment support.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1-5, in the embodiment of the present invention: an intelligent scoring device for a chemical carbon dioxide preparation experiment comprises a transparent glassware 1, a black experiment table top 2 for reducing the reflection degree, a fisheye lens 3 for expanding the video visual field, a deep learning neural network model 4, a computer 5 for rigorous and accurate scoring logic judgment, and an equipment support 6 matched with the transparent glassware 1; the equipment support 6 is used for clamping and fixing the transparent glassware 1 to perform carbon dioxide experiment preparation on the black experiment table top 2, and different color ring labels 101 are arranged on the equipment support 6 and the transparent glassware 1 and are used for increasing the identification degree of the transparent glassware 1; the fisheye lens 3 comprises a front-view lens 301 and a top-view lens 302, the front-view lens 301 is arranged on the black experiment table top 2 in a right-to-right mode, and the top-view lens 302 is arranged above the black experiment table top 2 in a hanging mode through a hanging bracket 303; the deep learning neural network model 4 is placed on the black experimental table 2, and the computer 5 is placed above the deep learning neural network model 4.
In the above embodiment, the transparent glassware 1 includes a beaker, a glass rod, a test tube, a 90 ° teaching tube and a gas collection bottle, color ring labels 101 of different colors are embedded on the glass surfaces of different glassware, and the transparent glassware 1 corresponding to the color ring label 101 of each color is recorded into the deep learning neural network model 4, so that the characteristics of each transparent glassware 1 are very obvious under the recorded video, and thus, under the image detection and calculation method, each transparent glassware 1 can be accurately detected, and the accuracy of the subsequent scoring module is improved.
In the above embodiment, the black experiment table top 2 is made of a black light absorption material, which can effectively reduce the light reflection degree of the transparent glassware 1, so that the display of the transparent glassware 1 in a video can be clearer.
In the above embodiment, the fisheye lens 3 transmits the recorded picture to the deep learning neural network model 4, and the deep learning neural network model 4 compiles a corresponding fisheye correction algorithm, so that the picture with large distortion can be corrected to a normal picture, which is beneficial to the implementation of the following target detection algorithm.
In the above embodiment, deep learning is used as a new direction of machine learning, which is a big hand in many important problems in the field of artificial intelligence, and is also a light in the field of education, but for a transparent glassware 1 such as a chemical apparatus, deep learning is sometimes an unwarranted one, and in target detection, because of different object shapes, problems such as illumination influence and shielding can occur at any position on an image, so that the requirements on the accuracy and real-time performance of a detection frame are high, based on the difficulty, the deep learning frame based on the existing basis is modified for symptom administration, so that a deep learning convolutional neural network frame which can be used for the transparent glassware 1 is realized, under the frame, the network can be used for learning image characteristics, and each glassware including a large test tube and a small test tube can be recognized with high precision and high efficiency, a small-difference transparent glass 1 of a large beaker and a small beaker; the network framework of the deep learning neural network model 4 adopts an SSD framework as a baseline, a standard SSD adopts a VGG-16 framework as a backbone network, and the SSD starts from a conv 43 layer and further comprises an FC7 layer (converted into a conv layer) of the original VGG-16 network, and the last Full Connection (FC) layer of the network is cut off; then it adds several progressively smaller conv layers, i.e. at the last conv82, conv92, conv102 and conv112 for prediction;
based on the above description, the whole deep learning neural network model 4 is composed of three parts, namely a standard SSD network, a Lightweight Scratch Network (LSN), and a bidirectional network; firstly, sampling an input image, then generating LSN characteristics through Lightweight Serial Operation (LSO), wherein a bidirectional network consists of a bottom-up scheme and a top-down scheme, transmitting low/middle-level high-level semantic information in the network, using VGG-16 as a main body, then generating a low layer/middle layer characteristic representation by the LSN, and injecting the low layer/middle layer characteristic representation into the main body characteristics of a subsequent standard prediction layer so as to improve the performance; the resulting properties of the current and previous layers are then combined in a bottom-up fashion in a bi-directional network, the top-down scheme in a bi-directional network containing independent parallel connections, injecting high-level semantic information from a later layer of the network to a previous layer.
Referring to fig. 6, to better explain the embodiment of the invention, an intelligent scoring method for chemical carbon dioxide extraction experiment is also provided, which comprises the following steps:
the first step is as follows: placing the transparent glassware 1 with the color ring label 101 on a black experiment table top 2, standing a student in front of the experiment table, and completing corresponding experiment operation according to experiment requirements;
the second step is that: monitoring each operation flow of students by the cooperative work of the two cameras of the front-view lens 301 and the top-view lens 302 in the fisheye lens 3 for expanding the video field;
the third step: the corrected forward-looking video and the normal top-looking video are transmitted to a computer 5 together, the computer 5 firstly carries out data cleaning on the acquired video to obtain a better preprocessing effect, and the acquired video is specified to be in a specification which can be transmitted to a deep learning neural network model 4, wherein the data cleaning mainly comprises fisheye correction on the forward-looking video; before fisheye correction, the front-view lens 301 is calibrated to obtain the internal and external parameters of the camera, and then fisheye correction is carried out through a fisheye correction algorithm in the computer 5, so that a normal video picture is obtained;
the fourth step: after the video enters the deep learning neural network model 4, a video with a detection effect is returned, and meanwhile, two video stream files containing all video information are provided, wherein the two video stream files are processing data of a top view video and a front view video, and the processing data contain the identification detection result of the chemical equipment of each frame in the video;
the fifth step: and (4) performing logic relation judgment scoring according to the position relation of the transparent glassware 1 and the equipment support 6 by the video stream file obtained in the fourth step and entering a scoring algorithm of the computer 5.
Based on the above description, the following specific examples are also provided:
the large experiment for preparing the carbon dioxide is divided into two parts, wherein one part is a simple device for preparing the carbon dioxide; one part is to use the built device to prepare carbon dioxide; the first part is the study on the using capacity and the operating capacity of students on chemical instruments; the second part is to examine the basic knowledge of students on the experiment of making carbon dioxide.
The first part of the invention judges whether the student scores according to two evaluation points, namely the sequence of building chemical equipment by the student and the airtightness of the device for detecting whether the student detects the device; whether the student operates the standard or not is judged according to the two files obtained by the method and the posture estimation of the student.
The second part of the invention can judge whether the gas collecting method of the students is an upward air exhaust method or not, and also judge some operation specifications, and finally judge whether the students use burning battens to be close to the gas collecting bottle mouth to test the carbon dioxide gas or not.
For flame detection, the embodiment provides a set of flame detection methods, and aims at solving the problem that the flame color model is generally a certain color interval at present and cannot accurately represent the flame color characteristics, flame colors are analyzed in different spaces, the distribution rule of flame pixel values in the spaces is searched, and a more accurate color probability distribution model is established according to the analysis; in the background of the experimental environment, a foreground extraction method based on a search box is provided, the color and the motion characteristics of flame are fully utilized, and the inter-frame correlation of the color and the motion is integrated to accurately extract a flame area; flame characteristics based on image blocks are researched, and multi-dimensional characteristics such as flame color significance characteristics, spatial gradient characteristics, inter-frame gradient characteristics, flicker characteristics and flame centroid movement characteristics are fused for flame characterization and identification; and then, judging whether the student puts the burning wood bar at the gas collection bottle mouth or not by judging the distance threshold value between the flame and the gas collection bottle mouth, judging the operation details of some specific actions, and finally judging whether the operation of the student is correct or not by judging the integral action flow by using the recognition result of the chemical equipment acquired in the deep learning neural network model 4 and the gesture, posture and action data information of the student.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be able to cover the technical solutions and the inventive concepts of the present invention within the technical scope of the present invention.
Claims (7)
1. An intelligent scoring device for a chemical carbon dioxide preparation experiment is characterized by comprising a transparent glassware (1), a black experiment table top (2) for reducing the light reflection degree, a fisheye lens (3) for expanding the video visual field, a deep learning neural network model (4), a computer (5) for rigorous and accurate scoring logic judgment, and an equipment support (6) matched with the transparent glassware (1) for use; the equipment support (6) is used for clamping and fixing the transparent glassware (1) to perform carbon dioxide experiment preparation on a black experiment table top (2), and different color ring labels (101) are arranged on the equipment support (6) and the transparent glassware (1); the fisheye lens (3) comprises a front-view lens (301) and a top-view lens (302), the front-view lens (301) is installed on the black experiment table top (2) in a right-to-right mode, and the top-view lens (302) is installed above the black experiment table top (2) in a hanging mode through a hanging frame (303); the deep learning neural network model (4) is placed on the black experiment desktop (2) and is connected with the computer (5), and the computer (5) is placed above the deep learning neural network model (4).
2. The intelligent scoring device for chemical carbon dioxide extraction experiments as claimed in claim 1, wherein: the transparent glassware (1) comprises a beaker, a glass rod, a test tube, a 90-degree teaching tube and a gas collecting bottle, color ring labels (101) with different colors are embedded on the glass surfaces of different utensils, and the transparent glassware (1) corresponding to the color ring labels (101) with each color is recorded into the deep learning neural network model (4).
3. The intelligent scoring device for chemical carbon dioxide extraction experiments as claimed in claim 1, wherein: the black experiment table top (2) is made of a black light absorption material.
4. The intelligent scoring device for chemical carbon dioxide extraction experiments as claimed in claim 1, wherein: the fisheye lens (3) transmits the recorded pictures into the deep learning neural network model (4), and a corresponding fisheye correction algorithm is compiled in the deep learning neural network model (4).
5. The intelligent scoring device for chemical carbon dioxide extraction experiments as claimed in claim 1, wherein: the deep learning neural network model (4) is composed of a standard SSD network, a lightweight scratch network and a bidirectional network.
6. An intelligent scoring method for chemical carbon dioxide extraction experiments as claimed in claim 1, comprising the steps of:
s1: putting a transparent glass vessel (1) with a color ring label (101) on a black experiment table top (2), standing a student in front of the experiment table, and completing corresponding experiment operation according to experiment requirements;
s2: monitoring each operation flow of students by the cooperative work of a front-view lens (301) and a double-camera of a top-view lens (302) in a fisheye lens (3) for expanding the visual field of a video;
s3: the corrected forward-looking video and the normal top-looking video are transmitted to a computer (5) together, the computer (5) firstly cleans the acquired video to obtain a better preprocessing effect, and the acquired video is specified to be in a specification which can be transmitted to a deep learning neural network model (4), wherein the data cleaning mainly comprises fisheye correction of the forward-looking video;
s4: after the video enters the deep learning neural network model (4), a video with a detection effect is returned, and meanwhile, two video stream files containing all video information are provided, wherein the two video stream files are top-view video and front-view video processing data which contain the identification detection result of the chemical equipment of each frame in the video;
s5: and (4) entering a scoring algorithm of the computer (5) through the video stream file obtained in the S4, and carrying out logic relation judgment scoring according to the position relation of the transparent glassware (1) and the equipment bracket (6).
7. The intelligent scoring method for the chemical carbon dioxide extraction experiment as recited in claim 6, wherein in S3, before the fisheye correction, the front-view lens (301) is calibrated to obtain the camera internal and external parameters, and then the fisheye correction is performed through a fisheye correction algorithm in the computer (5), so as to obtain a normal video picture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010203120.7A CN111312022A (en) | 2020-03-20 | 2020-03-20 | Intelligent scoring device and method for chemical carbon dioxide preparation experiment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010203120.7A CN111312022A (en) | 2020-03-20 | 2020-03-20 | Intelligent scoring device and method for chemical carbon dioxide preparation experiment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111312022A true CN111312022A (en) | 2020-06-19 |
Family
ID=71153501
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010203120.7A Pending CN111312022A (en) | 2020-03-20 | 2020-03-20 | Intelligent scoring device and method for chemical carbon dioxide preparation experiment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111312022A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115331156A (en) * | 2022-10-17 | 2022-11-11 | 成都西交智汇大数据科技有限公司 | Carbon dioxide preparation experiment scoring method, device, equipment and readable storage medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202114050U (en) * | 2011-05-08 | 2012-01-18 | 蒋云福 | Laboratory hood with replaceable tabletop |
| CN104091505A (en) * | 2014-08-04 | 2014-10-08 | 韦忠平 | Multifunctional experimental reactor |
| US8873086B2 (en) * | 2010-01-29 | 2014-10-28 | Xerox Corporation | Methods and system for consumable validity verification in prepaid document processing devices |
| CN105376471A (en) * | 2015-12-21 | 2016-03-02 | 景好 | Panorama shooting system based on moving platform and method |
| CN106875797A (en) * | 2015-12-14 | 2017-06-20 | 重庆尚尚科技有限公司 | It is a kind of to test with the device for producing carbon dioxide |
| CN206613506U (en) * | 2017-02-10 | 2017-11-07 | 云南同方科技有限公司 | A kind of multi-functional beaker |
| CN109460863A (en) * | 2018-10-29 | 2019-03-12 | 中国石油大学(华东) | Equipment state prediction method based on deep learning |
| CN109727172A (en) * | 2019-03-18 | 2019-05-07 | 上海中科教育装备集团有限公司 | A kind of artificial intelligence machine study experimental skill points-scoring system |
| CN110223202A (en) * | 2018-03-01 | 2019-09-10 | 彼乐智慧科技（北京）有限公司 | A kind of method and system of teaching stage property identification and scoring |
| CN209530927U (en) * | 2018-07-25 | 2019-10-25 | 杭州海康威视数字技术股份有限公司 | A kind of experimental bench |
| KR102087889B1 (en) * | 2019-11-11 | 2020-03-11 | 손동우 | A deep-learning eduacation system using arrow images |
-
2020
- 2020-03-20 CN CN202010203120.7A patent/CN111312022A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8873086B2 (en) * | 2010-01-29 | 2014-10-28 | Xerox Corporation | Methods and system for consumable validity verification in prepaid document processing devices |
| CN202114050U (en) * | 2011-05-08 | 2012-01-18 | 蒋云福 | Laboratory hood with replaceable tabletop |
| CN104091505A (en) * | 2014-08-04 | 2014-10-08 | 韦忠平 | Multifunctional experimental reactor |
| CN106875797A (en) * | 2015-12-14 | 2017-06-20 | 重庆尚尚科技有限公司 | It is a kind of to test with the device for producing carbon dioxide |
| CN105376471A (en) * | 2015-12-21 | 2016-03-02 | 景好 | Panorama shooting system based on moving platform and method |
| CN206613506U (en) * | 2017-02-10 | 2017-11-07 | 云南同方科技有限公司 | A kind of multi-functional beaker |
| CN110223202A (en) * | 2018-03-01 | 2019-09-10 | 彼乐智慧科技（北京）有限公司 | A kind of method and system of teaching stage property identification and scoring |
| CN209530927U (en) * | 2018-07-25 | 2019-10-25 | 杭州海康威视数字技术股份有限公司 | A kind of experimental bench |
| CN109460863A (en) * | 2018-10-29 | 2019-03-12 | 中国石油大学(华东) | Equipment state prediction method based on deep learning |
| CN109727172A (en) * | 2019-03-18 | 2019-05-07 | 上海中科教育装备集团有限公司 | A kind of artificial intelligence machine study experimental skill points-scoring system |
| KR102087889B1 (en) * | 2019-11-11 | 2020-03-11 | 손동우 | A deep-learning eduacation system using arrow images |
Non-Patent Citations (1)
| Title |
|---|
| TIANCAI WANG等: "Learning Rich Features at High-Speed for Single-Shot Object Detection", 《RESEARCHGATE》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115331156A (en) * | 2022-10-17 | 2022-11-11 | 成都西交智汇大数据科技有限公司 | Carbon dioxide preparation experiment scoring method, device, equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110796018B (en) | Hand motion recognition method based on depth image and color image | |
| Naim et al. | Unsupervised alignment of natural language instructions with video segments | |
| CN111665941B (en) | Virtual experiment-oriented multi-mode semantic fusion human-computer interaction system and method | |
| CN102622916A (en) | Human body acupuncture point projection demonstration method and device | |
| CN104376315A (en) | Detection method based on computer image processing and mode recognition and application of detection method | |
| CN109472247A (en) | Face identification method based on the non-formula of deep learning | |
| CN112116840B (en) | Job correction method and system based on image recognition and intelligent terminal | |
| TWI747686B (en) | A defect detection method and a defect detection device | |
| Fink et al. | Lsfb-cont and lsfb-isol: Two new datasets for vision-based sign language recognition | |
| CN111312022A (en) | Intelligent scoring device and method for chemical carbon dioxide preparation experiment | |
| CN117238024A (en) | Gesture recognition method based on improved YOLOv5 | |
| CN108846853A (en) | A kind of teaching behavior analysis method and device based on target following and attitude detection | |
| Sanmitra et al. | Machine Learning Based Real Time Sign Language Detection | |
| Yang et al. | Scb-dataset3: A benchmark for detecting student classroom behavior | |
| Lai et al. | Research on head-up and down behavior computer detection by deep learning and artificial intelligence | |
| CN114970701A (en) | Multi-mode fusion-based classroom interaction analysis method and system | |
| Vahdani et al. | Recognizing american sign language nonmanual signal grammar errors in continuous videos | |
| CN116128681A (en) | Intelligent scoring system for oxygen preparation experiment of middle school | |
| CN116343343A (en) | Intelligent evaluation method for crane lifting command action based on cloud end architecture | |
| Chen et al. | Intelligent Recognition of Physical Education Teachers' Behaviors Using Kinect Sensors and Machine Learning. | |
| Wang et al. | Design of Intelligent Detection System for Smoking Based on Improved YOLOv4. | |
| CN115049986A (en) | Flame detection method and system based on improved YOLOv4 | |
| Liu et al. | Development of a fatigue detection and early warning system for crane operators: A preliminary study | |
| Draxler et al. | An Environment-Triggered Augmented-Reality Application for Learning Case Grammar | |
| CN114255509A (en) | Student supervises appurtenance based on OpenPose |
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: 20200619 |
|
| RJ01 | Rejection of invention patent application after publication |