CN112199982A - Intelligent home system based on deep learning - Google Patents
Intelligent home system based on deep learning Download PDFInfo
- Publication number
- CN112199982A CN112199982A CN202010636003.XA CN202010636003A CN112199982A CN 112199982 A CN112199982 A CN 112199982A CN 202010636003 A CN202010636003 A CN 202010636003A CN 112199982 A CN112199982 A CN 112199982A
- Authority
- CN
- China
- Prior art keywords
- subsystem
- data
- module
- behavior
- abnormal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic 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
Abstract
The invention discloses an intelligent home system based on deep learning. The system comprises a video recording subsystem, an information processing subsystem, an intelligent control subsystem, a behavior data analysis subsystem and an early warning subsystem. The video recording subsystem realizes uninterrupted recording of real-time conditions in a user room within 24 hours by using a high-definition camera and machine learning; the information processing subsystem is used for carrying out image data processing and characteristic value extraction on the recorded video data so as to further detect abnormal behavior data; the convolutional neural network CNN algorithm is adopted to detect abnormal behavior data, abnormal behaviors are divided according to abnormal grades, different early warning measures are adopted according to a preset abnormal threshold and the abnormal grades, and the safety of a user is guaranteed to a great extent; the system also has a behavior data analysis function, and the daily behavior data of the user is analyzed by utilizing technologies such as big data analysis technology, artificial intelligence, machine learning and the like, so that the system can serve different users more humanizedly.
Description
Technical Field
The invention belongs to the technical field of deep learning, and particularly relates to an intelligent home system based on deep learning.
Background
The development of modern science and technology drives the development of the intelligent equipment industry, and the intellectualization of residential home is an important development trend. At present, the society of China is changing day by day, and the life style of people gradually moves to the mental level from the basic life needs of clothes, food, live, walking and the like, namely, people pay more attention to the quality of life. In recent years, emerging smart home markets are as fierce and have combined with emerging technologies, functions such as intelligent light control, intelligent electrical appliance control and security monitoring systems are achieved, safe and comfortable living environments are provided for people, but at present, most smart homes do not really achieve intellectualization, and many problems exist:
(1) the intelligent home industry in China starts late, so the prior related technology is still undeveloped, the stability of the prior intelligent home control system product is not strong, the conditions of incompatibility of a mobile phone system and the like exist, and the actual effect is greatly different from the expected effect.
(2) Different enterprises have different product docking protocols, so that the phenomenon that smart homes of different brands cannot be compatible exists, and the use of the smart homes by consumers is greatly restricted.
(3) Traditional intelligent house is as monitored control system, can't accomplish 24 hours incessant control, mostly needs the manual action data to intervene the discernment, and not only the discernment speed is slow, and the error rate is high.
In view of this, the traditional smart home with single function, complex operation and poor compatibility obviously cannot meet the requirements of people on the smart home at present. Therefore, designing an efficient and intelligent smart home system is a real need for development of all industries. The intelligent home system based on deep learning provided by the invention utilizes a computer vision technology, an image intelligent analysis technology and an artificial intelligence technology to detect, analyze and understand sequence images in a monitoring video. And (4) intercepting images according to the video, and judging whether an abnormal behavior state occurs in the house by adopting an intelligent video analysis technology, thereby achieving the effect of real-time early warning.
Disclosure of Invention
The intelligent home system based on deep learning provided by the invention uses related technologies such as machine learning and artificial intelligence, realizes the functions diversification, convenient operation, stability and high efficiency and the like of the home system, and makes up the defects of the traditional intelligent home system.
The invention is realized by the following steps: an intelligent home system based on deep learning comprises a video recording subsystem, an information processing subsystem, a behavior data analysis subsystem, an intelligent control subsystem and an early warning subsystem. Wherein, the video recording subsystem is connected with the information processing subsystem; the information processing subsystem is connected with the intelligent control subsystem and the behavior data analysis subsystem; the intelligent control subsystem is connected with the early warning subsystem.
The video recording subsystem includes: the device comprises a video receiving module and a video data storage module.
The video receiving module uses a plurality of high-definition cameras to shoot the indoor conditions of the user in real time.
The video data storage module uses a 1TB external SD card to store video data.
The information processing subsystem includes: an image processing module, a feature extraction module and a behavior pre-judging module, wherein the subsystems of the image processing module, the feature extraction module and the behavior pre-judging module are mainly used for carrying out behavior recognition and judgment on received video data,
the image processing module adopts a color image enhancement algorithm to enhance the effect of the image; and improving the image data by adopting a data coding mode.
The feature extraction module adopts an LBP feature extraction algorithm to extract features of the data after image processing and establish an abnormal behavior feature database.
The behavior pre-judging module adopts a feature matching algorithm, carries out analysis and judgment according to known feature information, traverses an abnormal behavior feature database and determines the abnormal behavior to be detected.
The behavior data analysis subsystem performs big data analysis on the user behavior data by using a RapidMiner tool to obtain more valuable information.
The intelligent control subsystem includes: the system comprises an abnormal behavior receiving module, a data detection module, an abnormal behavior recording module and a data updating module.
And the abnormal behavior receiving module is used for receiving the abnormal behavior data to be detected uploaded by the behavior pre-judging module in the information processing subsystem.
The data detection module adopts a Convolutional Neural Network (CNN) algorithm to construct a SoftMax classification model, detects data and finally determines abnormal behavior data according to a behavior recognition technology.
And the abnormal behavior recording module is used for recording abnormal behavior data according to different abnormal grade classifications by the system.
And the data updating module extracts the characteristic value of the new abnormal behavior data and adds the extracted characteristic value to the abnormal behavior characteristic database to update the data.
And the early warning subsystem carries out early warning according to the danger level in the abnormal behavior recording module.
Compared with the traditional intelligent home system with single function, the intelligent home system based on deep learning provided by the invention has the following advantages:
(1) the recorded image data has high quality, and two coding modes are adopted to improve the image data and the data compression ratio, so that the transmission of the image data provides efficiency on the basis of not changing the bandwidth.
(2) The function is diversified, except can intelligent control domestic appliance, can also carry out 24 hours incessant control to the indoor, and the unusual action of real-time indoor carries out record and early warning, guarantee user's safety, and the while system possesses independently learning ability, the continuous upgrading of lift system.
(3) The system has high response speed, can basically realize real-time monitoring and real-time response, and can realize millisecond-level response time on sudden abnormal behaviors, thereby greatly reducing the loss of users caused by emergency.
Drawings
FIG. 1 is a schematic diagram of a system configuration according to an embodiment of the present invention;
FIG. 2 is a functional flow diagram of a data detection module according to an embodiment of the present invention.
The labels in the figure are: 1, a video recording subsystem, a 1-1 video receiving module and a 1-2 video data storage module; 2, an information processing subsystem, a 2-1 image processing module, a 2-2 characteristic extraction module and a 2-3 behavior prejudgment module; the system comprises a behavior data analysis subsystem 3, an intelligent control subsystem 4, an abnormal behavior receiving module 4-1, a data detection module 4-2, an abnormal behavior recording module 4-3, a data updating module 4-4 and an early warning subsystem 5.
Detailed Description
Example (b):
as shown in fig. 1, an intelligent home system based on deep learning includes a video recording subsystem 1, an information processing subsystem 2, a behavior data analysis subsystem 3, an intelligent control subsystem 4, and an early warning subsystem 5. The video recording subsystem 1 comprises a video receiving module 1-1 and a video data storage module 1-2; the information processing subsystem 2 comprises an image processing module 2-1, a feature extraction module 2-2 and a behavior pre-judgment module 2-3; the intelligent control subsystem 4 comprises an abnormal behavior receiving module 4-1, a data detection module 4-2, an abnormal behavior recording module 4-3 and a data updating module 4-4.
The video recording subsystem 1 is connected with the information processing subsystem 2; the information processing subsystem 2 is connected with the behavior data analysis subsystem 3 and the intelligent control subsystem 4; the intelligent control subsystem 4 is connected with the early warning subsystem 5. The behavior pre-judging module 2-3 in the information processing subsystem 2 is connected with the abnormal behavior receiving module 4-1 in the intelligent control subsystem 4; and an abnormal behavior recording module 4-3 in the intelligent control subsystem 4 is connected with the early warning subsystem 5.
The video recording subsystem 1 comprises a video receiving module 1-1 and a video data storage module 1-2.
The video receiving module 1-1 is composed of a plurality of cameras, and is distributed at each position in a home for shooting the condition of each position in real time.
The video data storage module 1-2 receives the data transmitted by the video receiving module 1-1 through USB interface connection or signal transmission, and stores the video data by using a 1TB external SD card.
The information processing subsystem 2 comprises an image processing module 2-1, a feature extraction module 2-2 and a behavior prejudgment module 2-3.
In order to avoid the situations of video image blurring, serious noise and the like caused by the interference of an external environment, the image processing module 2-1 adopts a color image enhancement algorithm as a key algorithm for image preprocessing, amplifies the difference between the characteristics of different objects by enhancing useful information in an image, so that the characteristics of a region of interest become clear, and the effect of enhancing the image is achieved; meanwhile, the image data is coded by adopting a Huffman Codes (Huffman Codes) mode so as to improve the compression ratio and the quality of the image data and achieve the purpose of improving the transmission rate of the image on the basis of not changing the bandwidth.
Wherein the Huffman coding principle is as follows: according to the optimal coding theorem, in variable length coding, a code with a short word length is coded for a signal symbol with a high occurrence probability, and a code with a long character is coded for an information symbol with a low occurrence probability, and if the code word length is arranged in the reverse order of the symbol occurrence probability, the average code sub-length is always smaller than that of any other symbol sequence arrangement mode. Huffman coding has been shown to have the properties of optimal variable length codes, with the shortest average code length, approaching the entropy value.
As shown in formula 1, the huffman coding firstly arranges m symbols in the X signal sources in the order of probability from large to small, and then combines the last two pieces of information with the minimum probability into one message, thereby gradually reducing the number of messages of the signal sources; meanwhile, rearranging according to the probability of the signal source symbols from large to small, and repeating continuously until the result in the formula 2 appears, the signal sources stop arranging; respectively assigning 1 and 0 to the combined messages, and correspondingly assigning 1 and 0 to the last two messages; finally, the Huffman optimal variable length code is formed, and the data compression ratio of the image is improved by utilizing the Huffman coding, so that the memory consumption is reduced, and the image data quality is improved.
The feature extraction module 2-2 adopts an LBP feature extraction algorithm to extract features of the data after image processing, and establishes an abnormal behavior feature database.
The LBP feature extraction algorithm takes a neighborhood central pixel as a threshold value, the gray values of 8 adjacent pixels are compared with the threshold value, if the values of surrounding pixels are larger than the value of the central pixel, the position of the pixel point is marked as 1, and if not, the position of the pixel point is 0. Thus, 8 points in the 3 × 3 neighborhood can be compared to generate 8-bit binary numbers (usually converted into decimal numbers, i.e. LBP codes, 256 types in total), i.e. obtaining the LBP value of the central pixel point in the neighborhood, and using this value to reflect the texture information of the region.
The behavior pre-judging module 2-3 adopts a feature matching algorithm to analyze and judge according to the known feature information in the database, traverse the abnormal behavior feature database and determine the abnormal behavior to be detected.
The behavior data analysis subsystem 3 utilizes a RapidMiner tool to carry out big data analysis on the user behavior data to obtain more valuable information.
The intelligent control subsystem 4 comprises an abnormal behavior receiving module 4-1, a data detection module 4-2, an abnormal behavior recording module 4-3 and a data updating module 4-4.
The abnormal behavior receiving module 4-1 is used for receiving the data to be detected in the behavior prejudging module 2-3 in the information subsystem 2.
As shown in fig. 2, the data detection module 4-2 first extracts the received abnormal behavior data features by using a convolutional neural network CNN algorithm to form a data training set; constructing a classification model by using SoftMax; then, according to a behavior recognition technology, carrying out data intelligent analysis and matching on the classification model to determine abnormal behaviors; and finally, calculating the size of the abnormal value, dividing the abnormal grade, and making different responses according to the set abnormal threshold.
The convolutional neural network CNN algorithm is one of neural networks and is often used for research in the field of image recognition, each layer of nodes in the network are locally connected, and weight parameters of each layer are shared, so that the network is more suitable for directly taking a two-dimensional image as the input of the network, the complex characteristic extraction and data reconstruction process in the traditional recognition algorithm is avoided, and the image characteristics are automatically extracted through convolutional layers, so that a good extraction effect is achieved.
And the abnormal behavior recording module 4-3 is used for recording the recognized abnormal behavior data in real time according to different abnormal grade classifications by the system.
And the data updating module 4-4 extracts characteristic values from abnormal behavior data which are not in the database by utilizing a machine learning technology, and updates the characteristic values to the abnormal behavior characteristic database.
The early warning subsystem 5 can send early warning notices to users or public security organs or community properties in time according to the abnormal behavior data in the abnormal behavior recording module 4-3 in the intelligent control subsystem 4, and take different early warning response measures according to different conditions, thereby ensuring the personal and property safety of the users to a great extent.
In conclusion, the intelligent home system based on deep learning provided by the invention designs the video recording subsystem, and realizes uninterrupted recording of real-time conditions in a user room within 24 hours by using a high-definition camera, machine learning and the like; designing an information processing subsystem, and carrying out image data processing and characteristic value extraction on the recorded video data so as to further detect abnormal behavior data; the convolutional neural network CNN algorithm is adopted to detect abnormal behavior data, abnormal behaviors are divided according to abnormal grades, different early warning measures are adopted according to a preset abnormal threshold and the abnormal grades, and the safety of a user is guaranteed to a great extent; the system also has a behavior data analysis function, and the daily behavior data of the user is analyzed by utilizing technologies such as big data analysis technology, artificial intelligence, machine learning and the like, so that the system can serve different users more humanizedly.
While the foregoing is directed to the preferred embodiment of the present invention, it is understood that the foregoing is illustrative only and is not to be construed as limiting the scope of the invention, as numerous changes and modifications will become apparent to those skilled in the art in light of the foregoing description.
Claims (1)
1. The utility model provides an intelligent home systems based on deep learning which characterized in that includes: the system comprises a video recording subsystem, an information processing subsystem, an intelligent control subsystem, a behavior data analysis subsystem and an early warning subsystem; wherein, the video recording subsystem is connected with the information processing subsystem; the information processing subsystem is connected with the intelligent control subsystem and the behavior data analysis subsystem; the intelligent control subsystem is connected with the early warning subsystem;
the video recording subsystem includes: the system comprises a video receiving module and a video data storage module;
the video receiving module uses a plurality of high-definition cameras to shoot the indoor conditions of the user in real time;
the video data storage module stores video data by using a 1TB external SD card;
the information processing subsystem includes: the system comprises an image processing module, a feature extraction module and a behavior pre-judgment module, wherein the image processing module, the feature extraction module and the behavior pre-judgment module are mainly used for performing behavior recognition and judgment on received video data;
the image processing module adopts a color image enhancement algorithm to enhance the effect of the image; improving image data by adopting a Huffman data coding mode, and coding data according to the following formula during Huffman coding:
the feature extraction module adopts an LBP feature extraction algorithm to extract features of the data after image processing and establish an abnormal behavior feature database;
the behavior pre-judging module adopts a feature matching algorithm, carries out analysis and judgment according to known feature information, traverses an abnormal behavior feature database and determines an abnormal behavior to be detected;
the LBP feature extraction algorithm takes a neighborhood central pixel as a threshold value, the gray values of 8 adjacent pixels are compared with the threshold value, if the values of surrounding pixels are greater than the value of the central pixel, the position of the pixel point is marked as 1, otherwise, the position of the pixel point is 0; thus, 8 points in the 3x3 neighborhood can generate 8-bit binary numbers through comparison, the binary numbers are usually converted into decimal numbers, namely LBP codes, the LBP values of the central pixel points of the neighborhood are obtained, and the LBP values are used for reflecting the texture information of the area;
the abnormal behavior receiving module receives the abnormal behavior data to be detected uploaded by the behavior pre-judging module in the information processing subsystem;
the data detection module adopts a Convolutional Neural Network (CNN) algorithm to construct a SoftMax classification model, detects data and finally determines abnormal behavior data according to a behavior recognition technology;
the abnormal behavior recording module is used for recording abnormal behavior data according to different abnormal grade classifications by the system;
the data updating module is used for extracting the characteristic value of the new abnormal behavior data and then adding the extracted characteristic value to the abnormal behavior characteristic database to update data;
the behavior data analysis subsystem performs big data analysis on the user behavior data by using a RapidMiner tool to obtain more valuable information;
and the early warning subsystem carries out early warning according to the danger level in the abnormal behavior recording module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010636003.XA CN112199982B (en) | 2020-07-03 | 2020-07-03 | Intelligent home system based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010636003.XA CN112199982B (en) | 2020-07-03 | 2020-07-03 | Intelligent home system based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112199982A true CN112199982A (en) | 2021-01-08 |
| CN112199982B CN112199982B (en) | 2022-06-17 |
Family
ID=74005652
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010636003.XA Active CN112199982B (en) | 2020-07-03 | 2020-07-03 | Intelligent home system based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112199982B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170220854A1 (en) * | 2016-01-29 | 2017-08-03 | Conduent Business Services, Llc | Temporal fusion of multimodal data from multiple data acquisition systems to automatically recognize and classify an action |
| CN107292230A (en) * | 2017-05-09 | 2017-10-24 | 华南理工大学 | Embedded finger vein identification method based on convolutional neural network and having counterfeit detection capability |
| CN108289248A (en) * | 2018-01-18 | 2018-07-17 | 福州瑞芯微电子股份有限公司 | A kind of deep learning video encoding/decoding method and device based on content forecast |
| CN108694408A (en) * | 2017-04-11 | 2018-10-23 | 西安邮电大学 | A kind of driving behavior recognition methods based on depth sparseness filtering convolutional neural networks |
| CN109376613A (en) * | 2018-09-29 | 2019-02-22 | 东莞中国科学院云计算产业技术创新与育成中心 | Video brainpower watch and control system based on big data and depth learning technology |
| CN109543513A (en) * | 2018-10-11 | 2019-03-29 | 平安科技（深圳）有限公司 | Method, apparatus, equipment and the storage medium that intelligent monitoring is handled in real time |
| CN110895861A (en) * | 2018-09-13 | 2020-03-20 | 杭州海康威视数字技术股份有限公司 | Abnormal behavior early warning method and device, monitoring equipment and storage medium |
-
2020
- 2020-07-03 CN CN202010636003.XA patent/CN112199982B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170220854A1 (en) * | 2016-01-29 | 2017-08-03 | Conduent Business Services, Llc | Temporal fusion of multimodal data from multiple data acquisition systems to automatically recognize and classify an action |
| CN108694408A (en) * | 2017-04-11 | 2018-10-23 | 西安邮电大学 | A kind of driving behavior recognition methods based on depth sparseness filtering convolutional neural networks |
| CN107292230A (en) * | 2017-05-09 | 2017-10-24 | 华南理工大学 | Embedded finger vein identification method based on convolutional neural network and having counterfeit detection capability |
| CN108289248A (en) * | 2018-01-18 | 2018-07-17 | 福州瑞芯微电子股份有限公司 | A kind of deep learning video encoding/decoding method and device based on content forecast |
| CN110895861A (en) * | 2018-09-13 | 2020-03-20 | 杭州海康威视数字技术股份有限公司 | Abnormal behavior early warning method and device, monitoring equipment and storage medium |
| CN109376613A (en) * | 2018-09-29 | 2019-02-22 | 东莞中国科学院云计算产业技术创新与育成中心 | Video brainpower watch and control system based on big data and depth learning technology |
| CN109543513A (en) * | 2018-10-11 | 2019-03-29 | 平安科技（深圳）有限公司 | Method, apparatus, equipment and the storage medium that intelligent monitoring is handled in real time |
Non-Patent Citations (4)
| Title |
|---|
| S. RAJARAMAN等: "Assessment of an ensemble of machine learning models toward abnormality detection in chest radiographs", 《2019 41ST ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY (EMBC)》 * |
| 徐杰等: "实时系统下LBP与CNN结合的人脸识别方法", 《黑龙江科技大学学报》 * |
| 李晓雨等: "基于改进残差网络的车型识别算法", 《信息技术与信息化》 * |
| 种文艳: "智能监控系统中多目标检测与跟踪技术的研究", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112199982B (en) | 2022-06-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Huang et al. | Detection algorithm of safety helmet wearing based on deep learning | |
| Chino et al. | Bowfire: detection of fire in still images by integrating pixel color and texture analysis | |
| CN110517329A (en) | A kind of deep learning method for compressing image based on semantic analysis | |
| CN104954791A (en) | Method for selecting key frame from wireless distributed video coding for mine in real time | |
| CN102469302B (en) | Background model learning system for lighting change adaptation utilized for video surveillance | |
| CN113114400B (en) | Signal frequency spectrum hole sensing method based on time sequence attention mechanism and LSTM model | |
| CN111597983B (en) | Method for realizing identification of generated false face image based on deep convolutional neural network | |
| CN112446244A (en) | Human body action recognition method, neural network training method, related device and equipment | |
| CN108345900A (en) | Pedestrian based on color and vein distribution characteristics recognition methods and its system again | |
| CN115471423A (en) | Point cloud denoising method based on generation countermeasure network and self-attention mechanism | |
| CN111523483B (en) | Chinese meal dish image recognition method and device | |
| CN110244854A (en) | A kind of artificial intelligence approach of multi-class eeg data identification | |
| CN109259528A (en) | A kind of home furnishings intelligent mirror based on recognition of face and skin quality detection | |
| CN112199982B (en) | Intelligent home system based on deep learning | |
| Shen et al. | A method for detecting abnormal data of network nodes based on convolutional neural network | |
| Taj-Eddin et al. | A new compression technique for surveillance videos: evaluation using new dataset | |
| Li et al. | 3d human skeleton data compression for action recognition | |
| CN114373145A (en) | Monitoring video scene classification method based on ORB algorithm and capable of obtaining key frames | |
| WO2022105117A1 (en) | Method and device for image quality assessment, computer device, and storage medium | |
| CN111754459B (en) | Dyeing fake image detection method based on statistical depth characteristics and electronic device | |
| CN104731784A (en) | Method, system and mobile terminal for visual search | |
| Li et al. | Image object detection algorithm based on improved Gaussian mixture model | |
| CN115620386A (en) | Joint loss gait feature identification method based on improved residual error network | |
| CN110929118A (en) | Network data processing method, equipment, device and medium | |
| CN113537330B (en) | Non-invasive load identification method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20210108 Assignee: GUANGXI CUIFA TECHNOLOGY CO.,LTD. Assignor: GUILIN University OF TECHNOLOGY Contract record no.: X2022450000049 Denomination of invention: A smart home system based on deep learning Granted publication date: 20220617 License type: Common License Record date: 20221118 |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20210108 Assignee: Guangxi Julian Information Technology Co.,Ltd. Assignor: GUILIN University OF TECHNOLOGY Contract record no.: X2022450000633 Denomination of invention: A Smart Home System Based on Deep Learning Granted publication date: 20220617 License type: Common License Record date: 20221230 |
|
| EE01 | Entry into force of recordation of patent licensing contract |