CN102254224A - Internet of things electric automobile charging station system based on image identification of rough set neural network - Google Patents
Internet of things electric automobile charging station system based on image identification of rough set neural network Download PDFInfo
- Publication number
- CN102254224A CN102254224A CN2011101878289A CN201110187828A CN102254224A CN 102254224 A CN102254224 A CN 102254224A CN 2011101878289 A CN2011101878289 A CN 2011101878289A CN 201110187828 A CN201110187828 A CN 201110187828A CN 102254224 A CN102254224 A CN 102254224A
- Authority
- CN
- China
- Prior art keywords
- image
- neural network
- rough set
- rough
- electric automobile
- 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
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses an Internet of things electric automobile charging station system based on image identification of a rough set neural network, wherein a rough set is introduced to a neural network fusion model to serve as a prepositional system of a BP (back propagation) network; on the basis of taking the information expressed by each image as a knowledge system, the indistinguishable relationship of the rough set, an approximate set and an attribute reduction concept are utilized to perform enhancement processing and edge detection for the image respectively. By utilizing the technical scheme of the invention, when the system is used for identifying passengers and vehicles, the rough set theory and the neutral network are organically combined into a new fusion model, the real-time performance of the system can be favorably improved and the fault-tolerant capability of the system can also be favorably enhanced.
Description
Technical field
The present invention relates to Internet of Things electric automobile charging pile system, especially a kind of Internet of Things electric automobile charging pile system of the image recognition based on rough-set neural network.
Background technology
The charging electric vehicle facility is at present mainly based on charging pile, and a general charging pile once can only be a charging electric vehicle.Secondly the charging pile floor area is less, can be arranged on existing parking lot, shopping square and other are convenient to the place that electric automobile is stopped, charging pile only provides single-phase 220 volts of AC power, need charge to electric automobile by the vehicle-mounted charge machine, because the vehicle-mounted charge acc power is less, so charging pile generally adopts the trickle charge mode.Government's officer's car, enterprise's commercial car, Demonstration Garden be with operation areas such as car fixed set relatively, then can in government centralized office work place or enterprise gathers near zone, electric automobile charging station is built in the fair central area.Because it is bigger to build place people, the vehicle flowrate of a large amount of charging piles in Public Parking such as parking lot, airport, railway station, hotel, hospital, shopping center, supermarket or the parking stall, sub-district, set up camera and embedded processing systems in the electric automobile charging pile system based on Internet of Things, the using rough set algorithm is in the Pedestrians and vehicles target detection, image processing system can be discerned the Pedestrians and vehicles information that enters the charging station zone, and identifying information is built significant to the peaceful community that present China advances.But this current Flame Image Process and camera system just simply have audio and video information video recording memory function, do not possess or the image processing function deficiency, need a kind of algorithm that can carry out enhancement process and rim detection to image.
Summary of the invention
The Internet of Things electric automobile charging pile system that the purpose of this invention is to provide a kind of image recognition based on rough-set neural network, be used for pedestrian/vehicle identification, the two organically combines as new Fusion Model with rough set theory and neural network, both help to improve the real-time of system, helped the fault-tolerant ability of enhanced system again.
For achieving the above object, the present invention is achieved through the following technical solutions:
A kind of Internet of Things electric automobile charging pile system of the image recognition based on rough-set neural network is incorporated into rough set in the neural network Fusion Model, as the front-end system of BP network; As a knowledge system, on this basis, utilize the notion of the relation of can not differentiating of rough set, approximate set and attribute reduction respectively image to be carried out enhancement process and rim detection the information of each width of cloth image expression.
Rough set is used two big classes in Flame Image Process: a class is no decision analysis, and utilization can not be differentiated relation and image segmentation, enhancement process and cluster analysis are carried out in the value yojan; Another kind of is that decision analysis is arranged, and mainly comprises characteristics of image, as edge extracting, and to the pre-service of raw image data, as image classification.
The beneficial effect of the rough set image recognition of Internet of Things electric automobile charging pile of the present invention system: because neural network internal calculation complexity, the training time is long, and is easy to generate vibration.Be the accelerating network learning process, improve system real time, the present invention is incorporated into rough set in the neural network Fusion Model, as the front-end system of BP network.Use rough set in advance sample space to be handled, not only can eliminate redundant attributes, reduce the dimension of sample space, and can reduce the complicacy that makes up the rear end neural network, accelerate the speed of convergence of network, avoid producing " over-fitting " phenomenon.In addition, the random noise that neural network is brought in to a certain extent can filtering parameter sampling process, and rough set is very sensitive to noise.Both respectively have relative merits.Be used for pedestrian/vehicle identification, the two is organically being combined as new Fusion Model, both helping to improve the real-time of system, helping the fault-tolerant ability of enhanced system again.
Description of drawings
With embodiment the present invention is described in further detail with reference to the accompanying drawings below.
Fig. 1 is that the described algorithm of the embodiment of the invention is through the vehicle image behind the medium filtering;
Fig. 2 is that the embodiment of the invention is described through the differentiated vehicle image of background.
Embodiment
(1) rough set theory is in Application in Image Processing
In Flame Image Process, the application of rough set mainly contains two big classes: a class is not have the analysis of decision-making, mainly is to utilize relation can not differentiated and value yojan to carry out image segmentation, enhancement process and cluster analysis etc.; Another kind of is the analysis that decision-making is arranged, and mainly comprises characteristics of image, as edge extracting etc., also relates to the pre-service to raw image data, as image classification etc.Its basic thought be information with each width of cloth image expression as a knowledge system, on this basis, utilize attribute reduction in the rough set, can not differentiate relation and approximate collective concept carries out enhancement process and rim detection to image respectively.
(2) utilize the neural network recognition image
Neural network has the characteristics of self-organization, self study to information processing.Bond strength in the neural network between each neuron represents that with the weights size this weights can be given in advance, also can adapt to surrounding environment and constantly variation, and this process is called neuronic learning process.Neural network is used for image recognition and has the following advantages: the information distribution of neural network is stored in and links the weights coefficient, make network have very high fault-tolerance, and often have the partial loss of noise or input picture in the image recognition, so neural network can solve problem of image recognition preferably.In addition, the self-organization of neural network and self-learning function have loosened the required constraint condition of traditional images recognition methods greatly, make it demonstrate great superiority to problem of image recognition.The present invention adopts BP (feedforward of information, Error Feedback) three-layer network to make up model of cognition, and it comprises an input layer, a hidden layer and an output layer.According to recognition effect each time, the network weight that upgrades in time is till network reaches the precision that we preset.
(3) rough set and neural network fusion are used for pedestrian/vehicle identification
The present invention intends adopting feature extraction, rough set and BP neural network to merge the identification that realizes pedestrian/vehicle.Feature commonly used in the pedestrian detection has external appearance characteristic, motion feature and abstract characteristics, and abstract characteristics both can be used for describing external appearance characteristic, also can be used for describing motion feature.External appearance characteristic is used for static informations such as pedestrian's shape, size, texture of presentation video.Because the image static information is various, be not that each information all is identification pedestrian's necessary information, carry out attribute reduction so the pedestrian is discerned knowledge space by the introducing rough set theory, reduce the dimension of knowledge space, improve the reliability of judging the pedestrian.Variation and rule that motion feature produced when being used to describe the pedestrian and moving.Analyze the rigidity of movable body and periodically by the remaining light stream of calculating the moving region, have higher average remaining light stream, the pedestrian is distinguished according to the vehicle movement of nonrigid pedestrian's motion specific rigidity.Because there is the bad limitation that overcomes in single external appearance characteristic and motion feature, consider that a large amount of candidate regions and complicated scene can make recognition speed reduce, be difficult only with good sorter of resemblance training, therefore proposed resemblance and motion feature are merged, form a feature set, train classification with neural network.
Illustrate, vehicle flowrate is discerned, calculated to the vehicle that sails into.At first the image of taking is carried out gray scale and handle, the gray level image that the approximate characteristic up and down of process medium filtering and rough set is enhanced, as shown in Figure 1.Extract the background of image by the adaptive background extracting method, present image and background image subtract each other (background subtraction partial image) and draw as shown in Figure 2 black and white two-value contour images.Extract characteristics of image then, form proper vector, input BP neural network with the contrast of image template coupling, draws the neural network that trains.Utilize this neural network at last, new images is carried out vehicle identification.
Fig. 2 is a training sample, and white portion is an identifying object, can extract the features such as length breadth ratio, area occupied of white portion (area-of-interest, doubtful vehicle ' s contour), and input BP neural network can tentatively judge whether to be vehicle or large tracts of land noise.
Claims (2)
1. the Internet of Things electric automobile charging pile system based on the image recognition of rough-set neural network is characterized in that, rough set is incorporated in the neural network Fusion Model, as the front-end system of BP network; As a knowledge system, on this basis, utilize the notion of the relation of can not differentiating of rough set, approximate set and attribute reduction respectively image to be carried out enhancement process and rim detection the information of each width of cloth image expression.
2. the Internet of Things electric automobile charging pile system of the image recognition based on rough-set neural network according to claim 1, it is characterized in that: rough set is used two big classes in Flame Image Process: a class is no decision analysis, and utilization can not be differentiated relation and image segmentation, enhancement process and cluster analysis are carried out in the value yojan; Another kind of is that decision analysis is arranged, and mainly comprises characteristics of image, as edge extracting, and to the pre-service of raw image data, as image classification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011101878289A CN102254224A (en) | 2011-07-06 | 2011-07-06 | Internet of things electric automobile charging station system based on image identification of rough set neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011101878289A CN102254224A (en) | 2011-07-06 | 2011-07-06 | Internet of things electric automobile charging station system based on image identification of rough set neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102254224A true CN102254224A (en) | 2011-11-23 |
Family
ID=44981472
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011101878289A Pending CN102254224A (en) | 2011-07-06 | 2011-07-06 | Internet of things electric automobile charging station system based on image identification of rough set neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102254224A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567093A (en) * | 2011-12-20 | 2012-07-11 | 广州粤嵌通信科技股份有限公司 | Berth type recognizing method applied in visual berth automatic guiding system |
| CN104318256A (en) * | 2014-11-17 | 2015-01-28 | 上海埃蒙特自动化系统有限公司 | Method for achieving automatic detection and classification of glass defects on basis of computer software |
| CN104598916A (en) * | 2014-09-11 | 2015-05-06 | 单勇 | Establishment method of train recognition system and train recognition method |
| CN105574517A (en) * | 2016-01-22 | 2016-05-11 | 孟玲 | Electric vehicle charging pile with stable tracking function |
| CN105741258A (en) * | 2014-12-09 | 2016-07-06 | 北京中船信息科技有限公司 | Hull component image segmentation method based on rough set and neural network |
| CN107403154A (en) * | 2017-07-20 | 2017-11-28 | 四川大学 | A kind of gait recognition method based on dynamic visual sensor |
| CN108098775A (en) * | 2017-12-26 | 2018-06-01 | 河南理工大学 | Fuel adding method, device and storage medium |
| CN108462876A (en) * | 2018-01-19 | 2018-08-28 | 福州瑞芯微电子股份有限公司 | A kind of video decoding optimization adjusting apparatus and method |
| CN109450942A (en) * | 2018-12-25 | 2019-03-08 | 北京戴纳实验科技有限公司 | A kind of safety detection method and its detection device of laboratory management system for internet of things |
| CN109709418A (en) * | 2018-12-21 | 2019-05-03 | 国网北京市电力公司 | A kind of detection method and device of electrically-charging equipment, storage medium and processor |
| CN109784294A (en) * | 2019-01-25 | 2019-05-21 | 中国科学院合肥物质科学研究院 | A kind of fructus lycii image recognition localization method based on rough set theory candidate frame selection technique |
| WO2021253939A1 (en) * | 2020-06-18 | 2021-12-23 | 南通大学 | Rough set-based neural network method for segmenting fundus retinal vascular image |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040076984A1 (en) * | 2000-12-07 | 2004-04-22 | Roland Eils | Expert system for classification and prediction of generic diseases, and for association of molecular genetic parameters with clinical parameters |
| CN101110106A (en) * | 2007-06-21 | 2008-01-23 | 上海交通大学 | Multiple sensor information amalgamation method combining rough set and neural network |
| CN101430309A (en) * | 2008-11-14 | 2009-05-13 | 西安建筑科技大学 | Environmental quality evaluation method based on rough set-RBF neural network |
-
2011
- 2011-07-06 CN CN2011101878289A patent/CN102254224A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040076984A1 (en) * | 2000-12-07 | 2004-04-22 | Roland Eils | Expert system for classification and prediction of generic diseases, and for association of molecular genetic parameters with clinical parameters |
| CN101110106A (en) * | 2007-06-21 | 2008-01-23 | 上海交通大学 | Multiple sensor information amalgamation method combining rough set and neural network |
| CN101430309A (en) * | 2008-11-14 | 2009-05-13 | 西安建筑科技大学 | Environmental quality evaluation method based on rough set-RBF neural network |
Non-Patent Citations (3)
| Title |
|---|
| 徐立中等: "粗糙集理论在图像增强中的应用", 《数据采集与处理》, no. 03, 30 September 1999 (1999-09-30), pages 307 - 310 * |
| 李劲等: "基于粗糙神经网络的交通优化控制模型", 《计算机工程与应用》, no. 19, 1 July 2008 (2008-07-01), pages 215 - 219 * |
| 王晨等: "一种基于粗糙集理论的神经网络图像分割方法", 《计算技术与自动化》, no. 03, 15 September 2009 (2009-09-15), pages 86 - 89 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567093A (en) * | 2011-12-20 | 2012-07-11 | 广州粤嵌通信科技股份有限公司 | Berth type recognizing method applied in visual berth automatic guiding system |
| CN104598916B (en) * | 2014-09-11 | 2019-03-26 | 单勇 | A kind of construction method and train knowledge method for distinguishing of train identifying system |
| CN104598916A (en) * | 2014-09-11 | 2015-05-06 | 单勇 | Establishment method of train recognition system and train recognition method |
| CN104318256A (en) * | 2014-11-17 | 2015-01-28 | 上海埃蒙特自动化系统有限公司 | Method for achieving automatic detection and classification of glass defects on basis of computer software |
| CN105741258A (en) * | 2014-12-09 | 2016-07-06 | 北京中船信息科技有限公司 | Hull component image segmentation method based on rough set and neural network |
| CN105574517A (en) * | 2016-01-22 | 2016-05-11 | 孟玲 | Electric vehicle charging pile with stable tracking function |
| CN107403154A (en) * | 2017-07-20 | 2017-11-28 | 四川大学 | A kind of gait recognition method based on dynamic visual sensor |
| CN108098775A (en) * | 2017-12-26 | 2018-06-01 | 河南理工大学 | Fuel adding method, device and storage medium |
| CN108462876A (en) * | 2018-01-19 | 2018-08-28 | 福州瑞芯微电子股份有限公司 | A kind of video decoding optimization adjusting apparatus and method |
| CN109709418A (en) * | 2018-12-21 | 2019-05-03 | 国网北京市电力公司 | A kind of detection method and device of electrically-charging equipment, storage medium and processor |
| CN109709418B (en) * | 2018-12-21 | 2021-05-07 | 国网北京市电力公司 | Detection method and device for charging facility, storage medium and processor |
| CN109450942A (en) * | 2018-12-25 | 2019-03-08 | 北京戴纳实验科技有限公司 | A kind of safety detection method and its detection device of laboratory management system for internet of things |
| CN109784294A (en) * | 2019-01-25 | 2019-05-21 | 中国科学院合肥物质科学研究院 | A kind of fructus lycii image recognition localization method based on rough set theory candidate frame selection technique |
| WO2021253939A1 (en) * | 2020-06-18 | 2021-12-23 | 南通大学 | Rough set-based neural network method for segmenting fundus retinal vascular image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102254224A (en) | Internet of things electric automobile charging station system based on image identification of rough set neural network | |
| Asvadi et al. | DepthCN: Vehicle detection using 3D-LIDAR and ConvNet | |
| CN109597087B (en) | Point cloud data-based 3D target detection method | |
| He et al. | Rail transit obstacle detection based on improved CNN | |
| CN106096531A (en) | A kind of traffic image polymorphic type vehicle checking method based on degree of depth study | |
| CN104463241A (en) | Vehicle type recognition method in intelligent transportation monitoring system | |
| CN109117788A (en) | A kind of public transport compartment crowding detection method merging ResNet and LSTM | |
| CN103942575A (en) | System and method for analyzing intelligent behaviors based on scenes and Markov logic network | |
| CN104299009A (en) | Plate number character recognition method based on multi-feature fusion | |
| CN114359876B (en) | Vehicle target identification method and storage medium | |
| CN107705256B (en) | Forward vehicle-mounted video image enhancement method based on contact network semantics | |
| CN106845387A (en) | Pedestrian detection method based on self study | |
| CN104268898A (en) | Method for detecting density of passengers in bus on basis of image analysis | |
| CN103794050A (en) | Real-time transport vehicle detecting and tracking method | |
| Zhang et al. | Road marking segmentation based on siamese attention module and maximum stable external region | |
| CN109858738A (en) | A kind of vehicle follow gallop state driver emotion behavioral characteristics extract and discrimination method | |
| Meng et al. | Vehicle detection: A review | |
| Zimmer et al. | Real-time and robust 3d object detection within road-side lidars using domain adaptation | |
| Cong et al. | Track obstacle detection algorithm based on YOLOv3 | |
| Gomez-Donoso et al. | Three-dimensional reconstruction using SFM for actual pedestrian classification | |
| Jiao | [Retracted] Intelligent Research Based on Deep Learning Recognition Method in Vehicle‐Road Cooperative Information Interaction System | |
| Shang et al. | Regenerative braking control strategy based on multi-source information fusion under environment perception | |
| Ye et al. | Neural network‐based semantic segmentation model for robot perception of driverless vision | |
| CN103679156A (en) | Automatic identification and tracking method for various kinds of moving objects | |
| Choi et al. | Efficient driving scene image creation using deep neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20111123 |