CN107832769A - Object is located at the visual identity method in environment - Google Patents
Object is located at the visual identity method in environment Download PDFInfo
- Publication number
- CN107832769A CN107832769A CN201711098942.8A CN201711098942A CN107832769A CN 107832769 A CN107832769 A CN 107832769A CN 201711098942 A CN201711098942 A CN 201711098942A CN 107832769 A CN107832769 A CN 107832769A
- Authority
- CN
- China
- Prior art keywords
- cutting unit
- environment
- feature
- image
- storehouse
- 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
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- 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
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention discloses object to be located at the visual identity method in environment, including structure object storehouse, and title, local feature and the contour feature of familiar object again are stored in object storehouse；Collection object is located at the image in environment；Obtain image cloud data；Image is split according to cloud data；According to the contour feature in object storehouse, first time identification is carried out to each cutting unit, obtains the classification belonging to object in each cutting unit；According to the local feature in object storehouse, second is carried out to each cutting unit and is identified, obtains the title of object in each cutting unit.The present invention is split according to cloud data to image, and each object is in each cutting unit in image, in identification, avoids mutual interference.
Description
Technical field
The present invention relates to a kind of object to be located at the visual identity method in environment, belongs to visual identity field.
Background technology
Visual identification technology is indispensable important component in Computer Vision Recognition technical system, to object
When being identified, collection is object and the image of its surrounding enviroment, interference of the surrounding enviroment to identification in identification process
It is very serious.
The content of the invention
In order to solve the above-mentioned technical problem, the invention provides object to be located at the visual identity method in environment.
In order to achieve the above object, the technical solution adopted in the present invention is：
Object is located at the visual identity method in environment, including,
Object storehouse is built, title, local feature and the contour feature of familiar object again are stored in object storehouse；
Collection object is located at the image in environment；
Obtain image cloud data；
Image is split according to cloud data；
According to the contour feature in object storehouse, first time identification is carried out to each cutting unit, obtains object institute in each cutting unit
The classification of category；
According to the local feature in object storehouse, second is carried out to each cutting unit and is identified, obtains object in each cutting unit
Title.
In object storehouse, including some groups of contour features, the corresponding object classification of every group of contour feature, each object type
There are multiple objects, the corresponding title of each object and one group of local feature in not.
Local feature includes yardstick and rotational invariants.
Identify for the first time：In each cutting unit, picture point, which is gathered together, forms an image outline, extracts image outline
Feature, it is compared with each group contour feature in object storehouse, it is special with one group of profile of image outline similarity highest
Object classification corresponding to sign, it is the classification in the cutting unit belonging to object.
Second of identification：The local feature in cutting unit is extracted, is compared with the local feature under corresponding classification, phase
The object names like corresponding to degree one group of local feature of highest, it is the title of thing in the cutting unit.
Before splitting to image, cloud data is filtered.
The beneficial effect that the present invention is reached：The present invention is split according to cloud data to image, each object in image
In each cutting unit, in identification, mutual interference is avoided.
Embodiment
The following examples are only intended to illustrate the technical solution of the present invention more clearly, and can not limit the present invention with this
Protection domain.
Object is located at the visual identity method in environment, comprises the following steps：
Step 1, object storehouse is built, title, local feature and the contour feature of familiar object again are stored in object storehouse.
In object storehouse, including some groups of contour features, the corresponding object classification of every group of contour feature, each object type
There are multiple objects in not, the corresponding title of each object and one group of local feature, local feature include yardstick and invariable rotary
Amount.
Step 2, collection object is located at the image in environment.
Step 3, image cloud data is obtained, cloud data is filtered.
Step 4, image is split according to cloud data.
Step 5, the contour feature in object storehouse, first time identification is carried out to each cutting unit, it is single obtains each segmentation
Classification in member belonging to object.
Identify for the first time：In each cutting unit, picture point, which is gathered together, forms an image outline, extracts image outline
Feature, it is compared with each group contour feature in object storehouse, it is special with one group of profile of image outline similarity highest
Object classification corresponding to sign, it is the classification in the cutting unit belonging to object.
Step 6, the local feature in object storehouse, second is carried out to each cutting unit and is identified, it is single to obtain each segmentation
The title of object in member.
Second of identification：The local feature in cutting unit is extracted, is compared with the local feature under corresponding classification, phase
The object names like corresponding to degree one group of local feature of highest, it is the title of thing in the cutting unit.
The above method is split according to cloud data to image, and each object is in each cutting unit in image, is identifying
When, avoid mutual interference.
Described above is only the preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, without departing from the technical principles of the invention, some improvement and deformation can also be made, these are improved and deformation
Also it should be regarded as protection scope of the present invention.
Claims (6)
1. object is located at the visual identity method in environment, it is characterised in that：Including,
Object storehouse is built, title, local feature and the contour feature of familiar object again are stored in object storehouse；
Collection object is located at the image in environment；
Obtain image cloud data；
Image is split according to cloud data；
According to the contour feature in object storehouse, first time identification is carried out to each cutting unit, obtains object institute in each cutting unit
The classification of category；
According to the local feature in object storehouse, second is carried out to each cutting unit and is identified, obtains object in each cutting unit
Title.
2. object according to claim 1 is located at the visual identity method in environment, it is characterised in that：In object storehouse,
Including some groups of contour features, every group of contour feature corresponds to an object classification, has multiple objects in each object classification, each
The corresponding title of object and one group of local feature.
3. object according to claim 2 is located at the visual identity method in environment, it is characterised in that：Local feature includes
Yardstick and rotational invariants.
4. object according to claim 2 is located at the visual identity method in environment, it is characterised in that：Identify for the first time：
In each cutting unit, picture point, which is gathered together, forms an image outline, the feature of image outline is extracted, by itself and object storehouse
In each group contour feature be compared, object classification corresponding with one group of contour feature of image outline similarity highest, i.e.,
For the classification belonging to object in the cutting unit.
5. object according to claim 2 is located at the visual identity method in environment, it is characterised in that：Second of identification：
The local feature in cutting unit is extracted, is compared with the local feature under corresponding classification, one group of similarity highest is local
Object names corresponding to feature, it is the title of thing in the cutting unit.
6. object according to claim 1 is located at the visual identity method in environment, it is characterised in that：Carried out to image
Before segmentation, cloud data is filtered.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711098942.8A CN107832769A (en) | 2017-11-09 | 2017-11-09 | Object is located at the visual identity method in environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711098942.8A CN107832769A (en) | 2017-11-09 | 2017-11-09 | Object is located at the visual identity method in environment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107832769A true CN107832769A (en) | 2018-03-23 |
Family
ID=61654996
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711098942.8A Pending CN107832769A (en) | 2017-11-09 | 2017-11-09 | Object is located at the visual identity method in environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107832769A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110361710A (en) * | 2018-04-09 | 2019-10-22 | 北京京东尚科信息技术有限公司 | Object identification method and device based on laser point cloud |
| WO2022041274A1 (en) * | 2020-08-31 | 2022-03-03 | 苏州铭冠软件科技有限公司 | Visual recognition method for object in stereoscopic warehouse environment |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104243787A (en) * | 2013-06-06 | 2014-12-24 | 华为技术有限公司 | Photographing method and equipment, and photo management method |
| CN105488044A (en) * | 2014-09-16 | 2016-04-13 | 华为技术有限公司 | Data processing method and device |
| CN105933607A (en) * | 2016-05-26 | 2016-09-07 | 维沃移动通信有限公司 | Photographing effect adjusting method of mobile terminal and mobile terminal |
| CN106228570A (en) * | 2016-07-08 | 2016-12-14 | 百度在线网络技术（北京）有限公司 | A kind of Truth data determines method and apparatus |
| CN106250881A (en) * | 2016-08-25 | 2016-12-21 | 深圳大学 | A kind of target identification method based on three dimensional point cloud and system |
| CN106372574A (en) * | 2016-08-22 | 2017-02-01 | 湖南晖龙股份有限公司 | ROS operation system-based robot object identification method |
| CN106529375A (en) * | 2015-09-11 | 2017-03-22 | 上海乐今通信技术有限公司 | Mobile terminal and object feature identification method for image of mobile terminal |
| CN106778514A (en) * | 2016-11-24 | 2017-05-31 | 努比亚技术有限公司 | A kind of method and device for identifying object |
| CN107316048A (en) * | 2017-05-03 | 2017-11-03 | 深圳市速腾聚创科技有限公司 | Point cloud classifications method and device |
-
2017
- 2017-11-09 CN CN201711098942.8A patent/CN107832769A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104243787A (en) * | 2013-06-06 | 2014-12-24 | 华为技术有限公司 | Photographing method and equipment, and photo management method |
| CN105488044A (en) * | 2014-09-16 | 2016-04-13 | 华为技术有限公司 | Data processing method and device |
| CN106529375A (en) * | 2015-09-11 | 2017-03-22 | 上海乐今通信技术有限公司 | Mobile terminal and object feature identification method for image of mobile terminal |
| CN105933607A (en) * | 2016-05-26 | 2016-09-07 | 维沃移动通信有限公司 | Photographing effect adjusting method of mobile terminal and mobile terminal |
| CN106228570A (en) * | 2016-07-08 | 2016-12-14 | 百度在线网络技术（北京）有限公司 | A kind of Truth data determines method and apparatus |
| CN106372574A (en) * | 2016-08-22 | 2017-02-01 | 湖南晖龙股份有限公司 | ROS operation system-based robot object identification method |
| CN106250881A (en) * | 2016-08-25 | 2016-12-21 | 深圳大学 | A kind of target identification method based on three dimensional point cloud and system |
| CN106778514A (en) * | 2016-11-24 | 2017-05-31 | 努比亚技术有限公司 | A kind of method and device for identifying object |
| CN107316048A (en) * | 2017-05-03 | 2017-11-03 | 深圳市速腾聚创科技有限公司 | Point cloud classifications method and device |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110361710A (en) * | 2018-04-09 | 2019-10-22 | 北京京东尚科信息技术有限公司 | Object identification method and device based on laser point cloud |
| WO2022041274A1 (en) * | 2020-08-31 | 2022-03-03 | 苏州铭冠软件科技有限公司 | Visual recognition method for object in stereoscopic warehouse environment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103955660B (en) | Method for recognizing batch two-dimension code images | |
| CN109918987B (en) | Video subtitle keyword identification method and device | |
| CN103049750B (en) | Character identifying method | |
| US8644606B2 (en) | Method for visual image detection | |
| EP3678047B1 (en) | Method and device for recognizing identity of human target | |
| CN103679147A (en) | Method and device for identifying model of mobile phone | |
| WO2015017796A3 (en) | Learning systems and methods | |
| Bhosale Swapnali et al. | Feature extraction using surf algorithm for object recognition | |
| CN104156700A (en) | Face image glass removal method based on mobile shape model and weighted interpolation method | |
| GB2605723A (en) | Video segmentation based on weighted knowledge graph | |
| CN107832769A (en) | Object is located at the visual identity method in environment | |
| CN103745207A (en) | Feature extraction method and device for human face identification | |
| Hong et al. | Automatic recognition of flowers through color and edge based contour detection | |
| Prakash et al. | An efficient technique for ear detection in 3D: Invariant to rotation and scale | |
| CN103488966A (en) | Intelligent mobile phone capable of identifying real-name ticket information | |
| CN107369162B (en) | Method and system for generating insulator candidate target area | |
| CN103678280A (en) | Translation task fragmentization method | |
| US10223587B2 (en) | Pairing of images of postal articles with descriptors of singularities of the gradient field | |
| CN111429376A (en) | High-efficiency digital image processing method with high-precision and low-precision integration | |
| JP2013164832A (en) | Object recognition device and method | |
| CN107122781B (en) | A kind of plant leaf identification method based on blade shape and edge feature | |
| CN104408616A (en) | Supermarket prepayment method based on three-dimensional face recognition | |
| CN108241868B (en) | Method and device for mapping objective similarity to subjective similarity of image | |
| CN102254173A (en) | Method for segmenting fingerprint image based on combination of k-means clustering and linear neighborhood propagation | |
| JP2014115737A (en) | Image dictionary generation device, image dictionary generation method and computer program |
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: 20180323 |
|
| RJ01 | Rejection of invention patent application after publication |