CN107016364A - A kind of express delivery location acquiring method based on image recognition - Google Patents
A kind of express delivery location acquiring method based on image recognition Download PDFInfo
- Publication number
- CN107016364A CN107016364A CN201710211927.3A CN201710211927A CN107016364A CN 107016364 A CN107016364 A CN 107016364A CN 201710211927 A CN201710211927 A CN 201710211927A CN 107016364 A CN107016364 A CN 107016364A
- Authority
- CN
- China
- Prior art keywords
- express delivery
- image
- mark
- vehicle
- target
- 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
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
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
- G06Q10/0833—Tracking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Entrepreneurship & Innovation (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Development Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of express delivery location acquiring method based on image recognition, using deep learning and training of the convolutional neural networks for express delivery mark and the number-plate number, express delivery sign image identification model and Car license recognition model are obtained respectively, then the monitoring image of the public video monitoring net in city is utilized, pass sequentially through and obtain the identification that two kinds of identification models carry out vehicle identification and the number-plate number, it is finally based on express delivery article and associates corresponding relation with vehicle, utilize the geographical position of target video monitor terminal in the public video monitoring net in city, realize the positioning for express delivery position, overcome the deficiencies in the prior art, substantially increase the accuracy and efficiency of positioning.
Description
Technical field
The present invention relates to a kind of express delivery location acquiring method based on image recognition, belong to framing technical field.
Background technology
With expanding economy, the frequency of people's shopping online also gradually increases, and at the same time people are conveyed through to express delivery
The information real-time update of journey requires also more and more higher.
Now the more extensive express delivery location acquiring method of application be broadly divided into two kinds it is a kind of be according on express waybill
Bar code is scanned acquisition express delivery position, and another is to be put into positioner inside courier packages to obtain the position of express delivery
Put.Both approaches all come with some shortcomings.Former approach can not accomplish real-time express delivery positioning, and the latter can accomplish to express delivery
Positioning in real time, but the cost spent in positioning is higher.
Beijing LOIT Technology Co., Ltd. in 2015 discloses a kind of express delivery supervising device, and this invention is a dress
Put, device includes:Memory module, information detecting module, message processing module and global position system GPS are identified, effect is to deposit
Store up the mark for the courier packages that the express delivery supervising device is monitored;Information detecting module function is to detect the environment letter of courier packages
Breath, obtains environment measuring information;GPS module function is:The geographical position of courier packages is detected, the geographical position of courier packages is obtained
Confidence ceases;Message processing module function is:Environment measuring information and the GPS that taken at regular intervals described information detection module is obtained
Obtained geographical location information, and environment measuring information and geographical location information are converted into what express delivery Surveillance center can recognize
It is sent to after form in the express delivery Surveillance center, the environment measuring information and geographical location information and carries described information processing
The mark for the courier packages that module is obtained from the mark memory module.Above-mentioned supervising device can realize the monitoring to express delivery.
This device can position the position of courier packages in real time by GPS module, can also monitor the state of courier packages and fast
Pass residing ambient conditions.But there is also a little problems:Expense is too high during this device includes multiple module uses, this
Device is powered such as the situation of express delivery long-time undelivered using battery, it may occur that battery electric quantity is not enough and can not carry out
The situation of monitoring.
A kind of logistics information tracting monitor is authorized within 2016, its main thought is that positioning is set in warehouse shelf
System, GPS locator is equipped with haulage vehicle and terminal, and the position where commodity can be entered by GPS locator
Monitoring camera, monitoring camera and logistics are equipped with row Real-time Feedback, and haulage vehicle, distribution vehicle and terminal
Server connection is controlled, Surveillance center is that may be viewed by haulage vehicle, the reality scene in distribution vehicle and terminal, it is ensured that
The safety of commodity.This system tentatively realizes express delivery tracking, and the technology that it is combined using GPS and video monitoring is completed
The function of being tracked to express delivery, but there is also a little deficiencies, can accomplish that error is accurate fixed within 10 meters in city using GPS
Position, if hundreds of meters can be reached in some section possible errors in rural.Camera is only installed in warehouse and only sees that express delivery is reached
Situation during warehouse can not see real-time condition of the express delivery in transportation.
A kind of express delivery tracking system by injuring the proposition of nine pocket Industrial Co., Ltd.s authorized for 2015, this system,
There are some control computers to be connected with the server provided with database, server connection internet, mobile terminal is provided with identity card
Part reading device has simultaneously plugged Mobile phone card, and mobile terminal is connected by 2G, 3G or 4G mobile network with server, and mobile terminal connects
Receive gps signal and simultaneously beam back server, the information transfer that the identity document reading device of mobile terminal is read and typing server
Database, server is networked with public security ID card No. Help Center.This method can only accomplish to express delivery reach warehouse or
The information of express delivery can be followed the trail of when in courier's hand, it is impossible to accomplish the position in real time to express delivery in the way of express transportation
It is tracked.And the position of express delivery can not be tracked when the equipment in courier's hand cannot connect to GPS.
The content of the invention
The technical problems to be solved by the invention are to provide a kind of express delivery location acquiring method based on image recognition, base
In the deep learning for express delivery mark and the number-plate number and training, by the monitoring image of the public video monitoring net in city, enter
Row Intelligent Recognition, realizes the positioning for express delivery position.
In order to solve the above-mentioned technical problem the present invention uses following technical scheme:The present invention devises one and is based on image recognition
Express delivery location acquiring method, based on the public video monitoring net in city, positioned for the position during express delivery transfer,
Wherein, initialization express delivery sign image identification model and Car license recognition model, and two identification models are loaded on into city first
In each Video Monitoring Terminal in the public video monitoring net in city, then respectively be directed to each express delivery article, by express delivery article with
The number-plate number of loaded express delivery vehicle is associated, and is constituted express delivery article and is associated corresponding relation with vehicle;Then the express delivery
Location acquiring method, including following steps are performed in real time:
All Video Monitoring Terminals in the public video monitoring net in step A. cities, respectively for obtained video image in real time, are adopted
Analyzed with express delivery sign image identification model, judge to whether there is the vehicle with target express delivery mark in video image,
It is the Video Monitoring Terminal corresponding to the video image to be defined as target video monitor terminal, and enter step B;Otherwise weigh
Step A is performed again;
Each target video monitor terminal of step B. is directed to the vehicle in its video image with target express delivery mark respectively, adopts
Car license recognition model is used, the number-plate number with target express delivery mark vehicle is obtained, subsequently into step C;
Step C. respectively be directed to each target video monitor terminal, according in target video monitor terminal video image have mesh
The number-plate number of express delivery mark vehicle is marked, corresponding relation is associated with vehicle with reference to express delivery article, by the target video monitor terminal
Geographical position, be used as the real time position in its video image with target express delivery mark onboard vehicles each express delivery articles.
It is used as a preferred technical solution of the present invention:The express delivery sign image identification model and Car license recognition model,
Convolutional neural networks are respectively adopted and are trained acquisition.
It is used as a preferred technical solution of the present invention:The use convolutional neural networks are trained acquisition express delivery mark
Image recognition model, comprises the following steps:
Step a1. obtains each width picture with target express delivery mark, default all angles, is used as each width express delivery mark sample
Picture, and enter step a2;
Each width express delivery mark samples pictures are separately converted to 1mdb form express delivery mark samples pictures by step a2., into step
a3;
Step a3. uses all 1mdb forms express delivery mark samples pictures, is trained for convolutional neural networks, obtains fast
Pass sign image identification model.
It is used as a preferred technical solution of the present invention:The Car license recognition model includes license plate image identification model and car
Trade mark code identification model, especially by following steps, acquisition is trained using convolutional neural networks;
Each width number-plate number training image in number-plate number training set of images is separately converted to 1mdb form cars by step b1.
Trade mark code samples pictures, and enter step b2;
Step b2. uses all 1mdb forms number-plate number samples pictures, is trained for convolutional neural networks, obtains car
Board image recognition model;
The numeral that step b3. is directed in number-plate number image is split, and is carried out for each the digital subgraph being partitioned into
Gray processing, updates each digital subgraph, subsequently into step b4;
Step b4. uses each digital subgraph, is trained for convolutional neural networks, obtains number-plate number identification model.
It is used as a preferred technical solution of the present invention:In the step B, each target video monitor terminal is directed to respectively
There is the vehicle of target express delivery mark, first license plate image identification model, which is obtained, has target express delivery mark vehicle in its video image
License plate image, then for license plate image, using number-plate number identification model, obtain the car with target express delivery mark vehicle
Trade mark code.
It is used as a preferred technical solution of the present invention:The convolutional neural networks include 5 convolutional layers, 5 ponds
Layer and 3 feature extraction layers.
A kind of express delivery location acquiring method based on image recognition of the present invention using above technical scheme with it is existing
Technology is compared, with following technique effect:The express delivery location acquiring method based on image recognition that the present invention is designed, using volume
Product neutral net obtains express delivery sign image identification model respectively for deep learning and the training of express delivery mark and the number-plate number
With Car license recognition model, then using the monitoring image of the public video monitoring net in city, pass sequentially through and obtain two kinds of identification models
The identification of vehicle identification and the number-plate number is carried out, express delivery article is finally based on and associates corresponding relation with vehicle, it is public using city
The geographical position of target video monitor terminal in video monitoring net, realizes the positioning for express delivery position, overcomes prior art
Deficiency, substantially increase the accuracy and efficiency of positioning.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the express delivery location acquiring method based on image recognition designed by the present invention;
Fig. 2 is each step signal of embodiment.
Embodiment
The embodiment of the present invention is described in further detail with reference to Figure of description.
As shown in figure 1, the present invention devises a kind of express delivery location acquiring method based on image recognition, it is public based on city
Video monitoring net, is positioned, among specific actual application, first initially for the position during express delivery transfer altogether
Change express delivery sign image identification model and Car license recognition model, here, express delivery sign image identification model and Car license recognition model,
Convolutional neural networks are respectively adopted and are trained acquisition, convolutional neural networks include 5 convolutional layers, 5 pond layers and 3 spies
Levy extract layer.
Every express delivery has a mark of oneself company, and can the hard brush of mark the carrier from company just or
On express delivery car.We can by the mark on express delivery car quickly identify the express delivery car that this is that company.So we
Express delivery sign image is trained with convolutional neural networks, can be accomplished after final training to the fast of express delivery sign image identification
Pass sign image identification model.Wherein, express delivery sign image identification model is trained acquisition using following steps:
Step a1. obtains each width picture with target express delivery mark, default all angles, is used as each width express delivery mark sample
Picture, and enter step a2.Express delivery sign image may come from each express company, with the wind, justify logical etc., we are prior
Need with the express company corresponding to it contact by each width picture defining.
Each width express delivery mark samples pictures are separately converted to 1mdb form express delivery mark samples pictures by step a2., are entered
Step a3.
Step a3. uses all 1mdb forms express delivery mark samples pictures, is trained, obtains for convolutional neural networks
Obtain express delivery sign image identification model;Wherein, picture is divided into some pieces by convolutional neural networks by several 7*7 small lattice,
Then small lattice are carried out handling simultaneously;There are 5 convolutional layers, 5 pond layers and three features to carry layer in each small lattice;When data warp
The image that 27*27 is generated after first convolutional layer convolution and the optimization of pond layer is crossed, feature map numbers therein are 256;Through
Cross after second layer convolution and the optimization of pond layer, be changed into 13*13 image, wherein feature map numbers are 256;By the 3rd
After layer and the 4th layer of convolution and pond layer, it is changed into 13*13 image, wherein feature map numbers are 384;By last
After layer convolutional layer and pond layer, 6*6 image is generated, wherein feature map are 256;Next three feature extractions are entered
Layer draws the characteristic vector of 1024 dimensions;Last layer draws disaggregated model, as express delivery sign image to the training of 1024 dimensional vectors
Identification model.
Car license recognition model includes license plate image identification model and number-plate number identification model, especially by following steps,
Acquisition is trained using convolutional neural networks;
Each width number-plate number training image in number-plate number training set of images is separately converted to 1mdb form cars by step b1.
Trade mark code samples pictures, and enter step b2.
Step b2. uses all 1mdb forms number-plate number samples pictures, is obtained using with express delivery sign image identification model
Identical learning training method is obtained, is trained for convolutional neural networks, license plate image identification model is obtained.
The numeral that step b3. is directed in number-plate number image is split, and each digital subgraph for being partitioned into
Gray processing is carried out, shade is reduced to influence digital on car plate, each digital subgraph is updated, subsequently into step b4.
Step b4. uses each digital subgraph, is trained for convolutional neural networks, obtains number-plate number identification
Model.
License plate image identification model in express delivery sign image identification model and Car license recognition model, the number-plate number are recognized
Model is loaded in each Video Monitoring Terminal in the public video monitoring net in city, and each express delivery article is then directed to respectively,
Express delivery article and the number-plate number of loaded express delivery vehicle are associated, express delivery article is constituted and associates corresponding relation with vehicle;
Wherein, the corresponding numbering of each express delivery vehicle correspondence one, numbering and express company, the number-plate number are corresponded.It is complete in express delivery
Into after sorting, express delivery article is loaded which express delivery vehicle does a record by Quick Response Code barcode scanning gun, and count each express delivery
The express delivery article number of packages that vehicle is filled.Just completed every express delivery article work(corresponding with express delivery vehicle by the two steps
Can, quantity has been counted in entrucking, it is convenient when express delivery article is arrived to check whether express mail is lost.
In specific express delivery location acquiring method, practical application, including following steps are performed in real time:
All Video Monitoring Terminals in the public video monitoring net in step A. cities, respectively for obtained video image in real time, are adopted
Analyzed with express delivery sign image identification model, judge to whether there is the vehicle with target express delivery mark in video image,
It is that the Video Monitoring Terminal corresponding to the video image is defined as target video monitor terminal, and into step B, such as Fig. 2
Shown in middle a and b;Otherwise step A is repeated.
Each target video monitor terminal of step B. is directed to the car in its video image with target express delivery mark respectively
, first license plate image identification model identifies the license plate image with target express delivery mark vehicle, as shown in c in Fig. 2, and passes through
The function that ROI interest region is chosen in opencv intercepts out license plate image, then for license plate image, is recognized using the number-plate number
Model, obtains the number-plate number with target express delivery mark vehicle, as shown in d in Fig. 2, subsequently into step C.
Step C. is directed to each target video monitor terminal respectively, has according in target video monitor terminal video image
There is the number-plate number of target express delivery mark vehicle, corresponding relation is associated with vehicle with reference to express delivery article, the target video is monitored
The geographical position of terminal, is used as the real-time position in its video image with each express delivery article of target express delivery mark onboard vehicles
Put.
The express delivery location acquiring method based on image recognition designed by above-mentioned technical proposal, using convolutional neural networks
For the deep learning and training of express delivery mark and the number-plate number, express delivery sign image identification model and Car license recognition are obtained respectively
Model, then using the monitoring image of the public video monitoring net in city, passes sequentially through and obtains two kinds of identification models progress vehicle knowledges
Not with the identification of the number-plate number, it is finally based on express delivery article and associates corresponding relation with vehicle, utilizes the public video monitoring net in city
The geographical position of middle target video monitor terminal, realizes the positioning for express delivery position, overcomes the deficiencies in the prior art, significantly
Improve the accuracy and efficiency of positioning.
Embodiments of the present invention are explained in detail above in conjunction with accompanying drawing, but the present invention is not limited to above-mentioned implementation
Mode, can also be on the premise of present inventive concept not be departed from the knowledge that those of ordinary skill in the art possess
Make a variety of changes.
Claims (6)
1. a kind of express delivery location acquiring method based on image recognition, it is characterised in that:Based on the public video monitoring net in city,
Positioned for the position during express delivery transfer, wherein, express delivery sign image identification model is initialized first to be known with car plate
Other model, and two identification models are loaded in each Video Monitoring Terminal in the public video monitoring net in city, then
Each express delivery article is directed to respectively, express delivery article and the number-plate number of loaded express delivery vehicle are associated, and constitutes express delivery thing
Product associate corresponding relation with vehicle;Then the express delivery location acquiring method, including perform following steps in real time:
All Video Monitoring Terminals in the public video monitoring net in step A. cities, respectively for obtained video image in real time, are adopted
Analyzed with express delivery sign image identification model, judge to whether there is the vehicle with target express delivery mark in video image,
It is the Video Monitoring Terminal corresponding to the video image to be defined as target video monitor terminal, and enter step B;Otherwise weigh
Step A is performed again;
Each target video monitor terminal of step B. is directed to the vehicle in its video image with target express delivery mark respectively, adopts
Car license recognition model is used, the number-plate number with target express delivery mark vehicle is obtained, subsequently into step C;
Step C. respectively be directed to each target video monitor terminal, according in target video monitor terminal video image have mesh
The number-plate number of express delivery mark vehicle is marked, corresponding relation is associated with vehicle with reference to express delivery article, by the target video monitor terminal
Geographical position, be used as the real time position in its video image with target express delivery mark onboard vehicles each express delivery articles.
2. a kind of express delivery location acquiring method based on image recognition according to claim 1, it is characterised in that:It is described fast
Sign image identification model and Car license recognition model are passed, convolutional neural networks are respectively adopted and are trained acquisition.
3. a kind of express delivery location acquiring method based on image recognition according to claim 2, it is characterised in that:It is described to adopt
Acquisition express delivery sign image identification model is trained with convolutional neural networks, is comprised the following steps:
Step a1. obtains each width picture with target express delivery mark, default all angles, is used as each width express delivery mark sample
Picture, and enter step a2;
Each width express delivery mark samples pictures are separately converted to 1mdb form express delivery mark samples pictures by step a2., into step
a3;
Step a3. uses all 1mdb forms express delivery mark samples pictures, is trained for convolutional neural networks, obtains fast
Pass sign image identification model.
4. a kind of express delivery location acquiring method based on image recognition according to claim 2, it is characterised in that:The car
Board identification model includes license plate image identification model and number-plate number identification model, especially by following steps, using convolution god
Acquisition is trained through network;
Each width number-plate number training image in number-plate number training set of images is separately converted to 1mdb form cars by step b1.
Trade mark code samples pictures, and enter step b2;
Step b2. uses all 1mdb forms number-plate number samples pictures, is trained for convolutional neural networks, obtains car
Board image recognition model;
The numeral that step b3. is directed in number-plate number image is split, and is carried out for each the digital subgraph being partitioned into
Gray processing, updates each digital subgraph, subsequently into step b4;
Step b4. uses each digital subgraph, is trained for convolutional neural networks, obtains number-plate number identification model.
5. a kind of express delivery location acquiring method based on image recognition according to claim 4, it is characterised in that:The step
In rapid B, each target video monitor terminal is directed to the vehicle in its video image with target express delivery mark, first car plate figure respectively
As identification model obtains the license plate image with target express delivery mark vehicle, then for license plate image, known using the number-plate number
Other model, obtains the number-plate number with target express delivery mark vehicle.
6. a kind of express delivery location acquiring method based on image recognition according to any one in claim 2 to 5, it is special
Levy and be:The convolutional neural networks include 5 convolutional layers, 5 pond layers and 3 feature extraction layers.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710211927.3A CN107016364A (en) | 2017-04-01 | 2017-04-01 | A kind of express delivery location acquiring method based on image recognition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710211927.3A CN107016364A (en) | 2017-04-01 | 2017-04-01 | A kind of express delivery location acquiring method based on image recognition |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107016364A true CN107016364A (en) | 2017-08-04 |
Family
ID=59445714
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710211927.3A Pending CN107016364A (en) | 2017-04-01 | 2017-04-01 | A kind of express delivery location acquiring method based on image recognition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107016364A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108230498A (en) * | 2017-12-21 | 2018-06-29 | 合肥天之通电子商务有限公司 | A kind of Household security system for identifying courier |
CN110110950A (en) * | 2018-02-01 | 2019-08-09 | 北京京东振世信息技术有限公司 | Generate the method, apparatus and computer readable storage medium in dispatching road area |
CN111709294A (en) * | 2020-05-18 | 2020-09-25 | 杭州电子科技大学 | Express delivery personnel identity identification method based on multi-feature information |
CN112802027A (en) * | 2019-11-13 | 2021-05-14 | 成都天府新区光启未来技术研究院 | Target object analysis method, storage medium and electronic device |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130069765A1 (en) * | 2009-05-08 | 2013-03-21 | Citysync, Ltd. | Security systems |
CN105373782A (en) * | 2015-11-16 | 2016-03-02 | 深圳市哈工大交通电子技术有限公司 | Method of automatically recognizing hazardous chemical vehicle from image or video |
-
2017
- 2017-04-01 CN CN201710211927.3A patent/CN107016364A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130069765A1 (en) * | 2009-05-08 | 2013-03-21 | Citysync, Ltd. | Security systems |
CN105373782A (en) * | 2015-11-16 | 2016-03-02 | 深圳市哈工大交通电子技术有限公司 | Method of automatically recognizing hazardous chemical vehicle from image or video |
Non-Patent Citations (2)
Title |
---|
周凯龙: "基于深度学习的图像识别应用研究", 《基于深度学习的图像识别应用研究》 * |
焦蓬斐: "基于TLD的目标跟踪改进算法研究", 《基于TLD的目标跟踪改进算法研究》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108230498A (en) * | 2017-12-21 | 2018-06-29 | 合肥天之通电子商务有限公司 | A kind of Household security system for identifying courier |
CN110110950A (en) * | 2018-02-01 | 2019-08-09 | 北京京东振世信息技术有限公司 | Generate the method, apparatus and computer readable storage medium in dispatching road area |
CN112802027A (en) * | 2019-11-13 | 2021-05-14 | 成都天府新区光启未来技术研究院 | Target object analysis method, storage medium and electronic device |
CN111709294A (en) * | 2020-05-18 | 2020-09-25 | 杭州电子科技大学 | Express delivery personnel identity identification method based on multi-feature information |
CN111709294B (en) * | 2020-05-18 | 2023-07-14 | 杭州电子科技大学 | Express delivery personnel identity recognition method based on multi-feature information |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11935169B2 (en) | Displaying items of interest in an augmented reality environment | |
CN107016364A (en) | A kind of express delivery location acquiring method based on image recognition | |
CN109241349B (en) | Monitoring video multi-target classification retrieval method and system based on deep learning | |
CN106767812B (en) | A kind of indoor semantic map updating method and system based on Semantic features extraction | |
US10192190B2 (en) | Concepts for electronic door hangers | |
US11803803B2 (en) | Electronically connectable packaging systems configured for shipping items | |
CN107533691A (en) | Event and notice based on home | |
CN111553355B (en) | Monitoring video-based method for detecting and notifying store outgoing business and managing store owner | |
CN109255286A (en) | A kind of quick detection recognition method of unmanned plane optics based on YOLO deep learning network frame | |
US11462005B1 (en) | Image partitioning for re-identification | |
CN104794519A (en) | Cloud identification system and cloud identification method | |
CN110223310A (en) | A kind of line-structured light center line and cabinet edge detection method based on deep learning | |
CN115265545A (en) | Map matching navigation method, device, equipment and storage medium based on decision analysis | |
CN114022837A (en) | Station left article detection method and device, electronic equipment and storage medium | |
CN109934147B (en) | Target detection method, system and device based on deep neural network | |
CN102063726A (en) | Moving target classification method and system | |
CN111178464A (en) | Application of OCR recognition based on neural network in logistics industry express bill | |
CN115331147A (en) | Road spray detection method and system based on machine vision | |
CN112633116B (en) | Method for intelligently analyzing PDF graphics context | |
CN113454649B (en) | Target detection method, apparatus, electronic device, and computer-readable storage medium | |
Miyazato et al. | Development of a free viewpoint pedestrian recognition system using deep learning for multipurpose flying drone | |
CN111401104B (en) | Classification model training method, classification method, device, equipment and storage medium | |
Putranto et al. | Identification of safe landing areas with semantic segmentation and contour detection for delivery uav | |
CN111458699B (en) | Laser data verification method and device based on laser mileage calculation | |
Ji et al. | A Novel Vision-Based Truck-Lifting Accident Detection Method for Truck-Lifting Prevention System in Container Terminal |
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: 20170804 |
|
RJ01 | Rejection of invention patent application after publication |