CN110285752A - Specified object match cognization and localization method in a kind of image - Google Patents
Specified object match cognization and localization method in a kind of image Download PDFInfo
- Publication number
- CN110285752A CN110285752A CN201910449926.1A CN201910449926A CN110285752A CN 110285752 A CN110285752 A CN 110285752A CN 201910449926 A CN201910449926 A CN 201910449926A CN 110285752 A CN110285752 A CN 110285752A
- Authority
- CN
- China
- Prior art keywords
- sideline
- measurand
- pixel dimensions
- image
- localization method
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The match cognization and localization method of specified object in a kind of image, comprising the following steps: S100 measures this hand-held intelligent terminal starting shooting function, measurand and object of reference are placed in terminal screen middle section；S200 adjusts the relative position of measurand and object of reference；S300, gauger shoot image to measurand adjusted and object of reference；S400 calculates the Pixel Dimensions of complete edge line；S500 calculates the Pixel Dimensions in incomplete sideline；S600 calculates separately sideline proportionality coefficient according to complete sideline, incomplete sideline Pixel Dimensions and actual size；S700 carries out arithmetic mean to sideline proportionality coefficient and obtains proportionality coefficient；S800 calculates actual size according to sideline proportionality coefficient and measurand measurement Pixel Dimensions.For in existing human body and object measurement practical application, object of reference, measurand accurately can not be in same plane, object of reference fails the object of reference identification under the completely scenes such as display and match orientation problem, the image pixel dimensions of object of reference can be effectively calculated, to calculate the actual size of measurand.
Description
Technical field
The invention belongs to the untouched image location fields based on object of reference, and in particular to specified object in a kind of image
Match cognization and localization method.
Background technique
In daily life and modern industrial production, for the dimensional measurement of human body or object, traditional survey both can be used
The method that amount tool or equipment is directly contacted with measurand may be based on the light and handy portable intelligence such as smart phone, tablet computer
Energy terminal, carries out non-contact measurement.Because traditional measuring tool or equipment is inconvenient to carry, part scene, which is difficult to touch, to be reached,
It can not measure at any time；And it uses portable intelligent terminal (hereinafter referred to as " mobile phone terminal " or " terminal "), be based on
It shoots the processes such as image, identifying processing, correction measurement and carries out non-contact measurement, using more convenient and efficient.
It, can not since unknown photographer's distance is measured the distance of object in the contact type measurement based on image taking
Know the corresponding actual size of the Pixel Dimensions indicated in image with pixel unit, generallys use the object of reference based on size and calculate
The relative scale relationship of actual size and Pixel Dimensions.
Currently, being carried out based on intelligent terminals such as smart phone, tablet computers using non-contacting shooting image mode
Human body or object relative dimensions fields of measurement, have relevant open source literature in image.For example, a kind of mobile phone based on bank card
Measure length method (application number: 201510152845.7), it is a kind of based on currency length measurement method (application number:
201210391787) it, is proposed on the basis of the bank card of known dimensions, bank note respectively, primary standard substance and measurand is placed in system
One plane carries out image taking, and the size of actual object is calculated according to the pixel of image.
The above method when being measured using intelligent terminal, by by primary standard substance and testee be placed in unified plane into
Row measurement, and bank card is required independently to place, it can be explicitly shown the complete area of bank card, these requirements are surveyed in actual human body
It measures scene (such as be measured manpower and lift bank card) to be generally difficult to realize, needs to propose to adapt to the object of the actual scenes such as anthropological measuring
Body identification and matching locating method.
Summary of the invention
Present invention seek to address that can not be accurately in existing human body and object measurement application in object of reference, measurand
Same plane, object of reference fail the object of reference identification under the completely scenes such as display and match orientation problem, and are surveyed based on object of reference
Amount is as a result, calculate the actual size of measurand.In order to solve the above technical problems, the present invention adopts the following technical scheme that solution:
The match cognization and localization method of specified object in a kind of image, comprising the following steps:
Step S100: gauger's hand-held intelligent terminal keeps certain distance with measurand；Start image camera function,
Hand-held marker and measurand main body are placed in intelligent terminal to show in the middle part of screen；
Step S200: the relative position of adjustment measurand and marker makes measurand, hand-held marker
As far as possible close to same vertical plane；Marker allows to be at least partially obscured, but should at least show two angle points and three sides
Line.
Step S300: measurement people is to the measurand and object of reference progress image taking behind step S200 adjustment position.
Step S400: complete edge line Pixel Dimensions are calculated: according to the complete angle point in interface when taking pictures and sideline, calculated by mark
The Pixel Dimensions of quasi- object of reference；
Step S500: the Pixel Dimensions in incomplete sideline are calculated: completely shows that angle point, part are shown according to marker
The feature in sideline carries out local search, determines the start-stop position in incomplete sideline, and calculate the pixel for calculating incomplete sideline
Size.
Step S600: calculating the calculated result of sideline Pixel Dimensions according to sideline complete in step S400, step S500, point
It is not divided by obtain the proportionality coefficient of dimension conversion, sideline proportionality coefficient k with the actual size in sidelineiThe practical ruler in=object of reference sideline
Very little/object of reference sideline Pixel Dimensions；
Step S700: arithmetic mean is carried out to complete sideline, the calculated proportionality coefficient in incomplete sideline, is obtained final
Proportionality coefficient, proportionality coefficient
Step S800: according to the measurement Pixel Dimensions of proportionality coefficient k and measurand, calculate the practical ruler of measurand
It is very little；Measurand actual size=k × measurand Pixel Dimensions.
The technology of the present invention method is suitable for the untouched image location field based on object of reference, for existing human body and object
In bulk measurement practical application, object of reference, measurand accurately can not be in same plane, object of reference fails the completely scenes such as display
Under object of reference identification with match orientation problem, can effectively calculate the image pixel dimensions of object of reference, thus reckoning be tested pair
The actual size of elephant.
(1) it is measured based on non-contact method image taking in application, when object of reference part is blocked, object of reference and tested pair
As not in same vertical plane formula, using based on the opposite of vertical reference line, horizontal datum auxiliary measurand and object of reference
Position determines；
(2) side line length in the case of object of reference is blocked is determined using local feature searching algorithm.
Detailed description of the invention
Fig. 1 overall procedure
Fig. 2 intelligent terminal shoots central area and the measuring basis line of image
Fig. 3 marker angle point, complete/incomplete sideline schematic diagram
1- intelligent terminal shows screen, 2- vertical reference line, 3- horizontal datum, 4- screen center region, 5- object of reference
Complete sideline A, the incomplete sideline of 6- object of reference angle point A, 7- object of reference complete sideline B, 8- object of reference angle point B, 9- object of reference.
Specific embodiment
For apparent purpose of the present invention technical solution and advantage, with reference to embodiments further specifically to the present invention
It is bright.It should be appreciated that specific example described in this part is for explaining only the invention, it is not intended to limit the present invention.
The match cognization and localization method of specified object in a kind of image of the present invention, comprising the following steps:
Step S100: gauger's hand-held intelligent terminal keeps certain distance with measurand；Start image camera function,
The measurand main body of hand-held object of reference is placed in intelligent terminal to show in the middle part of screen, as shown in Figure 2；
In above-mentioned S100 step, measurement people is 3m-5m apart from representative value apart from measurand, meets practical most of fields
The application demand of scape；
In above-mentioned S100 step, hand-held object of reference refers to the object of actual size Ls known to bank card, identity card, bus card etc.
Body；
Step S200: the relative position of adjustment measurand and marker makes measurand, hand-held marker
As far as possible close to same plane；Marker allows to be at least partially obscured, but should at least show two angle points and three sidelines.
In above-mentioned steps S200, the sideline of object of reference allows part to be blocked, but at least must completely show on one side；
In above-mentioned steps S200, the sideline of object of reference should as far as possible with the vertical reference line or horizontal base in intelligent terminal screen
Directrix (as shown in Figure 2) keeping parallelism.
Step S300: measurement people is to the measurand and object of reference progress image taking behind only step S200 adjustment position.
Step S400: complete edge line Pixel Dimensions are calculated: according to the complete angle point in interface when taking pictures and sideline, calculated by mark
The Pixel Dimensions of quasi- object of reference；
Pixel Dimensions in above-mentioned steps S400, it is to be measured for measuring when referring to that intelligent terminal shooting image measures
The measurement unit of size length in the picture, i.e., the pixel quantity that size to be measured occupies.Following steps Pixel Dimensions are that this is fixed
Justice.
Complete sideline in above-mentioned steps S400 refers to the complete sideline in start-stop position, as shown in Figure 3.
In above-mentioned steps S400, the complete sideline of selection should as far as possible with intelligent terminal screen vertical reference line or horizontal base
Directrix keeping parallelism.
Step S500: the Pixel Dimensions in incomplete sideline are calculated: completely shows that angle point, part are shown according to marker
The feature in sideline carries out local search, determines the start-stop position in incomplete sideline, and calculate the pixel for calculating incomplete sideline
Size.
In above-mentioned steps S500, the incomplete sideline of object of reference refers to the sideline that part is blocked.
In above-mentioned steps S500, the start-stop position in incomplete sideline determines the line of angle point according to local search algorithm.
In above-mentioned steps S500, the incomplete sideline for calculating should be with the vertical reference line or level of intelligent terminal screen
Reference line is substantially parallel.
Step S600: calculating the calculated result of sideline Pixel Dimensions according to sideline complete in step S400, step S500, point
It is not divided by obtain the proportionality coefficient of dimension conversion, sideline proportionality coefficient k with the actual size in sidelineiThe practical ruler in=object of reference sideline
Very little/object of reference sideline Pixel Dimensions.
Step S700: arithmetic mean is carried out to complete sideline, the calculated proportionality coefficient in incomplete sideline, is obtained final
Proportionality coefficient, proportionality coefficient
Step S800: according to the measurement Pixel Dimensions of proportionality coefficient k and measurand, calculate the practical ruler of measurand
It is very little；Measurand actual size=k × measurand Pixel Dimensions.
The match cognization with localization method of specified object realize in above-mentioned image: (1) being clapped based on non-contact method image
Measurement is taken the photograph in application, when object of reference part is blocked, object of reference and measurand be not in same plane formula, using based on vertical base
The relative position of directrix, horizontal datum auxiliary measurand and object of reference determines；(2) it is determined using local feature searching algorithm
Object of reference be blocked in the case of side line length.
Claims (10)
1. the match cognization and localization method of specified object in a kind of image, comprising the following steps:
Step S100: gauger's hand-held intelligent terminal keeps certain distance with measurand；Start image camera function, by hand
It holds marker and measurand main body is placed in intelligent terminal and shows in the middle part of screen；
Step S200: the relative position of adjustment measurand and marker makes measurand, hand-held marker as far as possible
Close to same vertical plane；Marker allows to be at least partially obscured, but should at least show two angle points and three sidelines；
Step S300: measurement people is to the measurand and object of reference progress image taking behind step S200 adjustment position；
Step S400: complete edge line Pixel Dimensions are calculated: according to the complete angle point in interface when taking pictures and sideline, calculates tested standard ginseng
According to the Pixel Dimensions of object；
Step S500: the Pixel Dimensions in incomplete sideline are calculated: completely shows angle point, part display sideline according to marker
Feature carry out local search, determine the start-stop position in incomplete sideline, and calculate the Pixel Dimensions for calculating incomplete sideline；
Step S600: according to sideline complete in step S400, step S500 calculate sideline Pixel Dimensions calculated result, respectively with
The actual size in sideline is divided by obtain the proportionality coefficient of dimension conversion, sideline proportionality coefficient ki=object of reference sideline actual size/
Object of reference sideline Pixel Dimensions；
Step S700: arithmetic mean is carried out to complete sideline, the calculated proportionality coefficient in incomplete sideline, obtains final ratio
Coefficient, proportionality coefficient
Step S800: according to the measurement Pixel Dimensions of proportionality coefficient k and measurand, calculate the actual size of measurand；Quilt
Survey object actual size=k × measurand Pixel Dimensions.
2. the match cognization and localization method of specified object in a kind of image according to claim 1, it is characterised in that: on
It states in S100 step, measurement people is 3m-5m apart from representative value apart from measurand, to meet the application of practical most of scenes
Demand.
3. the match cognization and localization method of specified object in a kind of image according to claim 1, it is characterised in that: on
It states in S100 step, hand-held object of reference refers to the object of bank card, identity card, actual size Ls known to bus card.
4. the match cognization and localization method of specified object in a kind of image according to claim 1, it is characterised in that: on
It states in step S200, the sideline of object of reference allows part to be blocked, but at least must completely show on one side.
5. the match cognization and localization method of specified object in a kind of image according to claim 1, it is characterised in that: on
State in step S200, the sideline of object of reference should as far as possible in intelligent terminal screen vertical reference line or horizontal datum keep flat
Row.
6. the match cognization and localization method of specified object in a kind of image according to claim 1, it is characterised in that: on
The Pixel Dimensions in step S400 are stated, when referring to that intelligent terminal shooting image measures, for measuring size to be measured in image
The measurement unit of middle length, i.e., the pixel quantity that size to be measured occupies.
7. the match cognization and localization method of specified object in a kind of image according to claim 1, it is characterised in that: on
The complete sideline in step S400 is stated, refers to the complete sideline in start-stop position.
8. the match cognization and localization method of specified object in a kind of image according to claim 1, it is characterised in that: on
It states in step S400, the complete sideline of selection should keep flat with intelligent terminal screen vertical reference line or horizontal datum as far as possible
Row.
9. the match cognization and localization method of specified object in a kind of image according to claim 1, it is characterised in that: on
State in step S500, the incomplete sideline of object of reference refers to the sideline that is blocked of part, the start-stop position in incomplete sideline according to
Local search algorithm determines the line of angle point.
10. the match cognization and localization method of specified object in a kind of image according to claim 9, it is characterised in that:
In above-mentioned steps S500, the incomplete sideline for calculating should be with the vertical reference line or horizontal datum base of intelligent terminal screen
This keeping parallelism.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910449926.1A CN110285752B (en) | 2019-05-28 | 2019-05-28 | Method for matching, identifying and positioning specified object in image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910449926.1A CN110285752B (en) | 2019-05-28 | 2019-05-28 | Method for matching, identifying and positioning specified object in image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110285752A true CN110285752A (en) | 2019-09-27 |
| CN110285752B CN110285752B (en) | 2021-05-18 |
Family
ID=68002570
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910449926.1A Expired - Fee Related CN110285752B (en) | 2019-05-28 | 2019-05-28 | Method for matching, identifying and positioning specified object in image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110285752B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115496399A (en) * | 2022-10-12 | 2022-12-20 | 杭州余杭建筑设计院有限公司 | Unmanned aerial vehicle-based method and system for immediate updating and distributing foundation pit surveying tasks |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100147328A1 (en) * | 2008-12-15 | 2010-06-17 | Tokyo Electron Limited | Foreign matter removal method and storage medium |
| CN102102978A (en) * | 2009-12-16 | 2011-06-22 | Tcl集团股份有限公司 | Handheld terminal, and method and device for measuring object by using same |
| CN103630074A (en) * | 2013-11-29 | 2014-03-12 | 北京京东尚科信息技术有限公司 | Method and device for measuring minimum package volume of object |
| CN104359403A (en) * | 2014-11-21 | 2015-02-18 | 天津工业大学 | Plane part size measurement method based on sub-pixel edge algorithm |
| CN105928598A (en) * | 2016-04-20 | 2016-09-07 | 上海斐讯数据通信技术有限公司 | Method and system for measuring object mass based on photographing |
-
2019
- 2019-05-28 CN CN201910449926.1A patent/CN110285752B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100147328A1 (en) * | 2008-12-15 | 2010-06-17 | Tokyo Electron Limited | Foreign matter removal method and storage medium |
| CN102102978A (en) * | 2009-12-16 | 2011-06-22 | Tcl集团股份有限公司 | Handheld terminal, and method and device for measuring object by using same |
| CN103630074A (en) * | 2013-11-29 | 2014-03-12 | 北京京东尚科信息技术有限公司 | Method and device for measuring minimum package volume of object |
| CN104359403A (en) * | 2014-11-21 | 2015-02-18 | 天津工业大学 | Plane part size measurement method based on sub-pixel edge algorithm |
| CN105928598A (en) * | 2016-04-20 | 2016-09-07 | 上海斐讯数据通信技术有限公司 | Method and system for measuring object mass based on photographing |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115496399A (en) * | 2022-10-12 | 2022-12-20 | 杭州余杭建筑设计院有限公司 | Unmanned aerial vehicle-based method and system for immediate updating and distributing foundation pit surveying tasks |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110285752B (en) | 2021-05-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10083522B2 (en) | Image based measurement system | |
| US9482515B2 (en) | Stereoscopic measurement system and method | |
| US8326022B2 (en) | Stereoscopic measurement system and method | |
| US9667955B2 (en) | Method of calibrating a camera | |
| US9286506B2 (en) | Stereoscopic measurement system and method | |
| CN109341537A (en) | Dimension measurement method and device based on binocular vision | |
| EP4310446A1 (en) | Groove depth measurement method, apparatus and system, and laser measurement device | |
| EP2310799B1 (en) | Stereoscopic measurement system and method | |
| CN110285752A (en) | Specified object match cognization and localization method in a kind of image | |
| WO2021170051A1 (en) | Digital photogrammetry method, electronic device, and system | |
| KR101314101B1 (en) | System for three-dimensional measurement and method therefor | |
| KR101996226B1 (en) | Apparatus for measuring three-dimensional position of subject and method thereof | |
| US20240159621A1 (en) | Calibration method of a portable electronic device | |
| CN110455263A (en) | A kind of underground space smart phone image positioning method | |
| AU2009249003B2 (en) | Stereoscopic measurement system and method | |
| JP2002352231A (en) | Method and device for detecting end point of human body | |
| RU2471147C2 (en) | Stereoscopic measuring system and method | |
| CN105847679B (en) | Image acquisition method and electronic equipment | |
| CN108986025A (en) | Based on the high-precision different time image mosaic of incomplete appearance rail information and bearing calibration | |
| KR101745626B1 (en) | Apparatus and method for location-information | |
| CN111141218A (en) | Image ranging algorithm based on multiple small markers | |
| KR100597529B1 (en) | A method of acquiring 3-dimensional coordinates of short range multiple images without surveying reference points | |
| CN111397532A (en) | Three-dimensional measurement method based on marker | |
| TW200946862A (en) | Position-measuring apparatus and method thereof |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210518 |