CN112258471B - Rolling door state detection method and system - Google Patents
Rolling door state detection method and system Download PDFInfo
- Publication number
- CN112258471B CN112258471B CN202011127173.1A CN202011127173A CN112258471B CN 112258471 B CN112258471 B CN 112258471B CN 202011127173 A CN202011127173 A CN 202011127173A CN 112258471 B CN112258471 B CN 112258471B
- Authority
- CN
- China
- Prior art keywords
- line
- line feature
- door
- image
- state
- 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.)
- Active
Links
Images
Classifications
-
- 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
- G06T7/136—Segmentation; Edge detection involving thresholding
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a method and a system for detecting the state of a rolling door, relates to the technical field of computer image processing, and mainly solves the technical problem of how to detect the opening and closing state of the rolling door conveniently and accurately. The invention comprises the following steps: collecting an image of a rolling shutter door; carrying out projection transformation on the roller shutter door image by adopting a predetermined transformation matrix; performing LSD algorithm processing on the image after projection transformation to obtain an algorithm detection result; calculating line characteristics according to the algorithm detection result; filtering the line characteristics; performing line reconstruction on the filtered line characteristics; performing de-duplication on the line characteristics after line reconstruction; and judging the state of the roller shutter door based on the line characteristics after the weight removal. The invention is more efficient and accurate and has more robust performance. The detection result of the invention has lower requirements on the ambient light and shadow, and the camera which can accurately shoot the rolling shutter door can be directly connected without reinstalling equipment, so that the usability and convenience performance are more excellent. And the labor participation is not needed, so that the working pressure of store managers is reduced.
Description
Technical Field
The invention relates to the technical field of computer image processing, in particular to a rolling shutter door state detection method and system based on an LSD algorithm.
Background
Roller shutter doors are generally widely used in shops, and thus it is very important to detect the open/close state of the roller shutter door. In the prior art, there are generally 3 methods for detecting the opening and closing states of a rolling door:
1. the conventional method is to detect the door opening and closing state of the rolling shutter door based on a hardware sensor, for example, a door opening and closing reminding device and a door opening and closing reminding method, and the method is to detect the door opening and closing state based on a light sensor.
2: based on the detection of the manual reading, the opening and closing state of the rolling shutter door is observed through human eyes.
3: the detection of the door opening and closing state of the rolling shutter door based on the video image is in a blank stage, taking information acquired by a China network as an example, and similarly, the method is an elevator door opening and closing detection method (CN 106986248A) based on the camera image, and the method extracts an edge image by using an Otsu binarization algorithm and Canny edge detection, and extracts linear coordinate information by using Hough linear detection.
However, the existing detection method has the following disadvantages:
(1) For detection methods based on traditional hardware sensors, equipment needs to be reinstalled, the cost is high, and the detection method is affected by the environment to a certain extent.
(2) Based on the manual visual reading method, the user needs to participate in observing the video manually, which is extremely time-consuming and labor-consuming.
(3) Compared with the LSD algorithm, the straight line detection by adopting Hough transformation consumes more time, and the influence of the detection precision controlled parameters is larger.
Disclosure of Invention
One of the purposes of the invention is to provide a method and a system for detecting the state of a rolling door, which solve the technical problem of how to detect the opening and closing state of the rolling door conveniently and accurately in the prior art. Numerous advantageous effects can be achieved in the preferred embodiments of the present invention, as described in detail below.
In order to achieve the above purpose, the present invention provides the following technical solutions:
the invention discloses a rolling door state detection method, which comprises the following steps:
collecting an image of a rolling shutter door;
performing projection transformation on the rolling shutter door image by adopting a predetermined transformation matrix;
performing LSD algorithm processing on the image after projection transformation to obtain an algorithm detection result;
calculating line characteristics according to the algorithm detection result;
filtering the line feature;
performing line reconstruction on the filtered line characteristics;
performing de-duplication on the line characteristics after line reconstruction;
and judging the state of the roller shutter door based on the line characteristics after the weight removal.
Further, the method further comprises the following steps:
reading detection zone coordinates of a preset image detection zone;
and obtaining the transformation matrix through svd algorithm according to the detection area coordinates.
Further, the judging the state of the rolling door based on the line characteristics after the duplication removal includes:
performing line calculation on the line characteristics after the weight removal, and primarily judging the state of the roller shutter door according to the line quantity threshold value;
if the state of the roller shutter door is judged to be in a door closing state, calculating the line segment distance of the line characteristic judgment after the duplication removal;
and carrying out secondary judgment on the state of the rolling shutter door according to the interval threshold value.
Further, the line calculation is performed on the line characteristic judgment after the duplication removal, and the rolling door state is primarily judged according to the line quantity threshold, including:
if the line number of the line characteristics after the weight removal is smaller than the line number threshold value, the rolling door is in a door opening state; otherwise, the door is closed.
Further, the secondary judging of the state of the rolling door according to the interval threshold value includes:
if the line segment spacing of the line features after the weight removal is smaller than the spacing threshold value, the rolling shutter door is in a door closing state;
and if the line segment spacing of the line features after the weight removal is greater than or equal to the spacing threshold value, the rolling shutter door state is a door opening state.
Further, the filtering the line feature includes:
when the length of the roller shutter door image is not 0, deleting the line feature if the length of the line feature is calculated to be smaller than a length threshold value; otherwise, preserving the line feature;
when the gradient of the rolling door image is not 0, deleting the line feature if the gradient of the line feature is calculated to be smaller than a gradient threshold value; otherwise, preserving the line feature;
when the angle of the rolling door image is not 0, deleting the line feature if the angle of the line feature is calculated to be smaller than an angle threshold value; otherwise, the line feature is preserved.
Further, the de-duplicating the line characteristics after the line reconstruction includes:
if the starting point and the ending point of the current line feature are the same as those of the previous line feature, deleting the current line feature, and reserving the previous line feature.
Further, the method further comprises the following steps:
setting an image detection area, a length threshold value, a gradient threshold value, an angle threshold value, a distance threshold value and a line number threshold value of an LSD algorithm.
Further, the method further comprises the following steps:
initializing a memory space, a length threshold, a gradient threshold, an angle threshold, a pitch threshold and a line number threshold of an LSD algorithm.
The invention provides a state detection system for a configured rolling door, which is characterized by comprising the following components:
the acquisition module is used for acquiring the roller shutter door image;
the initialization module is used for carrying out projection transformation on the rolling shutter door image by adopting a predetermined transformation matrix;
the processing module is used for carrying out LSD algorithm processing on the image after projection transformation to obtain an algorithm detection result; calculating line characteristics according to the algorithm detection result; filtering the line feature; performing line reconstruction on the filtered line characteristics; performing de-duplication on the line characteristics after line reconstruction;
and the result acquisition module is used for judging the state of the roller shutter door based on the line characteristics after the weight removal.
The method and the system for detecting the state of the rolling door have the following beneficial technical effects:
firstly, collecting an image of a rolling door; then adopting a predetermined transformation matrix to carry out projection transformation on the rolling shutter door image; performing LSD algorithm processing on the image after projection transformation to obtain an algorithm detection result; then, calculating line characteristics according to the algorithm detection result; filtering the line feature; performing line reconstruction on the filtered line characteristics; performing de-duplication on the line characteristics after line reconstruction; and finally, judging the state of the roller shutter door based on the line characteristics after the weight removal. Therefore, compared with the Hough algorithm, the LSD algorithm adopted by the invention is more efficient and accurate, and related algorithm parameters are more robust with Hough linear detection. The detection result of the invention has lower requirements on the ambient light and shadow, and can be directly connected with the camera which can accurately shoot the rolling shutter door without reinstalling equipment, thereby having more excellent usability and convenience. And the labor participation is not needed, so that the working pressure of store management operators is reduced.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a rolling door status detection method of the present invention;
FIG. 2 is a schematic flow chart of another method for detecting the status of a rolling shutter door according to the present invention;
FIG. 3 is a flow chart of step S8 of the present invention;
fig. 4 is a schematic structural view of the rolling door state detection system of the present invention.
In the figure, a 1-acquisition module, a 2-initialization module, a 3-processing module and a 4-result acquisition module are adopted.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail below. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, based on the examples herein, which are within the scope of the invention as defined by the claims, will be within the scope of the invention as defined by the claims.
Referring to fig. 1 and 2, a rolling door state detection method of the present invention includes:
s1: collecting an image of a rolling shutter door;
s2: performing projection transformation on the rolling shutter door image by adopting a predetermined transformation matrix;
s3: performing LSD algorithm processing on the image after projection transformation to obtain an algorithm detection result;
s4: calculating line characteristics according to the algorithm detection result;
s5: filtering the line feature;
s6: performing line reconstruction on the filtered line characteristics;
s7: performing de-duplication on the line characteristics after line reconstruction;
s8: and judging the state of the roller shutter door based on the line characteristics after the weight removal.
It should be explained that the LSD algorithm (Line Segment Detector) is a straight line detection segmentation algorithm, which can obtain a detection result with sub-pixel level accuracy in a linear time.
Compared with the Hough algorithm, the LSD algorithm adopted by the invention is more efficient and accurate, and related algorithm parameters are more robust with Hough linear detection. The detection result of the invention has lower requirements on the ambient light and shadow, and can be directly connected with the camera which can accurately shoot the rolling shutter door without reinstalling equipment, thereby having more excellent usability and convenience. And the labor participation is not needed, so that the working pressure of store management operators is reduced.
The invention also includes:
setting an image detection area, a length threshold value, a gradient threshold value, an angle threshold value, a distance threshold value and a line number threshold value of an LSD algorithm.
It should be explained that, after the invention collects the rolling door image, the things on the rolling door image need to be screened, namely an image detection area is set; then, the set image check is determined and storedThe method comprises the steps of determining a region, wherein the method is to limit according to the size of an acquired rolling door image, the maximum value of the length and width coordinates of an image detection region is the maximum value of the length and width of the acquired rolling door image, and the minimum value is 0; finally, setting LSD algorithm parameters, and setting an optional length threshold value Thresh for the LSD algorithm len An optional gradient threshold Thresh grad An optional angle threshold Thresh ang Distance threshold Thresh spac Line number threshold Thresh num Etc.
The invention also includes:
initializing a memory space, a length threshold, a gradient threshold, an angle threshold, a pitch threshold and a line number threshold of an LSD algorithm.
It should be explained that, when the present invention is running in a computer, an initialization operation is performed after the image detection area is read or before the method of the present invention is started, and a normal operation is performed for the computer, so that the computer is in a normal state and the normal operation is ensured when the present invention is used.
The invention also includes:
reading detection zone coordinates of a preset image detection zone;
and obtaining the transformation matrix through svd algorithm according to the detection area coordinates.
It should be noted that,
s2, performing projection transformation on the rolling shutter door image by adopting a predetermined transformation matrix; the predetermined transformation matrix specifically comprises:
and calculating a projective transformation matrix of the coordinates of the read image detection area, wherein the formula is as follows:
wherein (x ', y ', w ') represents the coordinates of the image detection area after transformation, (x, y, w) represents the coordinates of the image detection area before transformation,/->
Representing the projective transformation matrix a that is the predetermined transformation matrix of the present invention. Sitting with image detection in AWhen the transformation matrix is obtained, firstly, the RoI coordinate, the Src coordinate and the Dst coordinate are generated through the coordinates of the input image detection area, and the coordinates are arranged according to the upper left, the upper right, the lower right and the lower left except the RoI coordinate, and the calculation formula is as follows:
the Src coordinates available are:
the available Dst coordinates are:
subsequently, the Src coordinates and the Dst coordinates are combined, and the projective transformation matrix a is solved by a svd algorithm. Wherein the SVD algorithm (Singular Value Decomposition) is a singular value decomposition algorithm, solving a projective transformation matrix.
Finally, the coordinates of the image detection area after projection transformation are obtained through the projection transformation matrix A and the coordinates of the image detection area.
Step S3, performing LSD algorithm processing on the image after projection transformation to obtain an algorithm detection result;
step S4, calculating line characteristics according to the algorithm detection result, wherein the step specifically comprises the following steps:
and calculating line characteristics of the algorithm detection result, wherein a length formula is as follows:
in which x is s ,y s Representing the coordinates of the start of the line feature, x e ,y e Representing line feature end point coordinates;
the angle formula is:
wherein alpha is an radian value obtained by an LSD detection algorithm;
for gradient calculation, a 3×3 gradient template is firstly applied to calculate an image to obtain a gradient matrix of the image, and then the starting point coordinates x are used s ,y s Endpoint coordinate x e ,y e Calculating the slope and intercept of the straight line of the line segment, wherein the formula is as follows:
secondly, the number judgment is carried out on the x direction or the y direction, and the formula is as follows:
subsequently, the principal data (more in the x-direction or y-direction), such as x, is iterated in combination with the linear equation s →x e Or y s →y e Wherein, combining the gradient matrix generated in the first step, generating a gradient value under the generated sub-pixel coordinate by using a bilinear interpolation method, and storing the gradient value to obtain a gradient value vector:
[g 0 g 1 ... g n ](n＝max(||x e -x s ||,||y e -y s ||))，
and taking an average value of the gradient matrix to generate a gradient characteristic value of the line characteristic, wherein the formula is as follows:
step S5, filtering the line feature, including:
when the length of the roller shutter door image is not 0, deleting the line feature if the length of the line feature is calculated to be smaller than a length threshold value; otherwise, preserving the line feature;
when the gradient of the rolling door image is not 0, deleting the line feature if the gradient of the line feature is calculated to be smaller than a gradient threshold value; otherwise, preserving the line feature;
when the angle of the rolling door image is not 0, deleting the line feature if the angle of the line feature is calculated to be smaller than an angle threshold value; otherwise, the line feature is preserved.
It should be explained that:
step S5, filtering the line characteristics, wherein the step specifically comprises the following steps:
if the length parameter of the image detection area is greater than 0, filtering the length of the line characteristic, wherein the formula is as follows:
middle del (f) i ) Representing the deletion of the ith line feature, save (f i ) Indicating retention of the ith line feature, len i Representing the length of the ith line feature;
if the gradient parameter of the image detection area is larger than 0, performing line characteristic gradient filtering, wherein the formula is as follows:
middle grad i A gradient representing an ith line feature;
if the angle parameter of the image detection area is larger than 0, line characteristic angle filtering is carried out, and the formula is as follows:
ang in middle i Representing the angle of the ith line feature.
Step S6, performing line reconstruction on the filtered line characteristics, wherein the method specifically comprises the following steps:
the invention aims at a rolling door image, wherein the rolling door is generally a uniformly distributed parallel line segment, so the line reconstruction changes an x coordinate corresponding to a y coordinate of a line characteristic starting point to 0, and an x coordinate corresponding to a y coordinate of an end point to an image detection area width RoI_w, and the formula is as follows:
original line characteristics:
f＝{(x s ,y s )，(x e ,y e )，len,ang,grad}
reconstruction of post-line features
f＝{(0,y s )，(RoI_w,y e )，len,ang,grad}。
Step S7, the performing de-duplication on the line characteristics after the line reconstruction includes:
if the starting point and the ending point of the current line feature are the same as those of the previous line feature, deleting the current line feature, and reserving the previous line feature.
It should be noted that,
step S7, performing de-duplication on the line characteristics after line reconstruction, wherein the method specifically comprises the following steps:
and removing the line characteristics after the line reconstruction, namely sorting the y coordinates of the starting points, iterating the line characteristics, deleting the current line characteristics if the current line characteristics are the same as the previous starting points and the previous finishing points, and reserving the previous line characteristics until all line characteristics are iterated.
Step S8, the step of determining the state of the rolling door based on the line characteristics after the duplication removal includes:
performing line calculation on the line characteristics after the weight removal, and primarily judging the state of the roller shutter door according to the line quantity threshold value;
if the state of the roller shutter door is judged to be in a door closing state, calculating the line segment distance of the line characteristic judgment after the duplication removal;
and carrying out secondary judgment on the state of the rolling shutter door according to the interval threshold value.
Preferably, the calculating the line of the line characteristic judgment after the duplication removal, and primarily judging the state of the roller shutter door according to the threshold value of the line quantity, includes:
if the line number of the line characteristics after the weight removal is smaller than the line number threshold value, the rolling door is in a door opening state; otherwise, the door is closed.
Preferably, the secondary judging of the state of the rolling shutter door according to the interval threshold value includes:
if the line segment spacing of the line features after the weight removal is smaller than the spacing threshold value, the rolling shutter door is in a door closing state;
and if the line segment spacing of the line features after the weight removal is greater than or equal to the spacing threshold value, the rolling shutter door state is a door opening state.
It should be noted that,
referring to fig. 3, in step S8, the step of determining the state of the rolling door based on the line characteristics after the duplication removal specifically includes:
and (3) checking the line quantity threshold value of the line characteristics after the duplication removal, if the line quantity threshold value is smaller than the threshold value, considering the door opening state, otherwise, entering the next step, wherein a checking formula is as follows:
where size (f) represents the number of line features.
By iterating the line characteristics, calculating the line segment distance by combining the starting point and the starting point of the current line characteristic with the starting point of the next line characteristic, and generating a distance array [ sp ] 0 ,sp 1 ,...,sp i-1 ]The calculation formula is as follows:
f in i →y s ，f i →y e The starting point y coordinate and the ending point y coordinate representing the ith line feature are averaged for distance.
Firstly removing the maximum value and the minimum value in the interval array, secondly calculating the average value in the interval array, taking the average value as the final interval value, and adopting the formula as follows:
and (3) checking the distance threshold value for the distance value, if the distance value is smaller than the threshold value, considering a door closing state, otherwise, judging the door opening state, wherein the formula is as follows:
referring to fig. 4, a configuration rolling door state detection system of the present invention includes:
the acquisition module 1 is used for acquiring images of the rolling shutter door;
an initialization module 2, configured to perform projective transformation on the rolling shutter door image by using a predetermined transformation matrix;
the processing module 3 is used for performing LSD algorithm processing on the image after projection transformation to obtain an algorithm detection result; calculating line characteristics according to the algorithm detection result; filtering the line feature; performing line reconstruction on the filtered line characteristics; performing de-duplication on the line characteristics after line reconstruction;
and a result acquisition module 4, configured to determine a state of the rolling door based on the line characteristics after the duplication removal.
It should be noted that the present invention further includes a configuration module, where the configuration module is configured to set a length threshold, a gradient threshold, an angle threshold, a pitch threshold, and a line number threshold of the image detection area and the LSD algorithm.
Compared with the detection method based on the traditional hardware sensor, the detection method does not need to reinstall equipment, has low cost and is not influenced by environment. Compared with a method based on manual visual reading, the method does not need to take part in observing the video manually, and is time-saving and labor-saving. Compared with the straight line detection by Hough transformation, the method has the advantages of high efficiency, high precision and high reliability.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (9)
1. A method of detecting a state of a roll-up door, comprising:
collecting an image of a rolling shutter door;
performing projection transformation on the rolling shutter door image by adopting a predetermined transformation matrix;
performing LSD algorithm processing on the image after projection transformation to obtain an algorithm detection result;
calculating line characteristics according to the algorithm detection result;
filtering the line feature;
performing line reconstruction on the filtered line characteristics;
performing de-duplication on the line characteristics after line reconstruction;
judging the state of the roller shutter door based on the line characteristics after the weight removal;
wherein the filtering the line feature comprises:
when the length of the roller shutter door image is not 0, deleting the line feature if the length of the line feature is calculated to be smaller than a length threshold value; otherwise, preserving the line feature;
when the gradient of the rolling door image is not 0, deleting the line feature if the gradient of the line feature is calculated to be smaller than a gradient threshold value; otherwise, preserving the line feature;
when the angle of the rolling door image is not 0, deleting the line feature if the angle of the line feature is calculated to be smaller than an angle threshold value; otherwise, preserving the line feature;
the line reconstruction of the filtered line features comprises the following steps:
changing an x coordinate corresponding to a y coordinate of the line feature starting point to 0, and changing an x coordinate corresponding to a y coordinate of the end point to an image detection area width RoI_w, wherein the formula is as follows:
original line characteristics:
f＝{(x s ,y s )，(x e ,y e )，len,ang,grad}
reconstruction post-line features:
f＝{(0,y s )，(RoI_w,y e )，len,ang,grad}
wherein x is s An x coordinate representing the line feature start point coordinate; y is s A y coordinate representing a line feature start coordinate; x is x e An x coordinate representing a line feature endpoint coordinate; y is e A y coordinate representing a line feature end point coordinate; roI_w represents the image detection area width; len represents the length of the line feature; ang represents the angle of the line feature; grad represents the gradient of the line feature.
2. The rolling door state detection method according to claim 1, further comprising:
reading detection zone coordinates of a preset image detection zone;
and obtaining the transformation matrix through svd algorithm according to the detection area coordinates.
3. The rolling door state detection method according to claim 1, wherein the judging the rolling door state based on the line characteristics after the weight removal includes:
performing line calculation on the line characteristics after the weight removal, and primarily judging the state of the roller shutter door according to the line quantity threshold value;
if the state of the roller shutter door is judged to be in a door closing state, calculating the line segment distance of the line characteristic judgment after the duplication removal;
and carrying out secondary judgment on the state of the rolling shutter door according to the interval threshold value.
4. A method for detecting a state of a rolling shutter door according to claim 3, wherein the performing line calculation on the line characteristic judgment after the weight removal, and preliminarily judging the state of the rolling shutter door according to a threshold value of the number of lines, comprises:
if the line number of the line characteristics after the weight removal is smaller than the line number threshold value, the rolling door is in a door opening state; otherwise, the door is closed.
5. A method for detecting a state of a rolling shutter door according to claim 3, wherein the performing the secondary judgment of the state of the rolling shutter door according to the pitch threshold value comprises:
if the line segment spacing of the line features after the weight removal is smaller than the spacing threshold value, the rolling shutter door is in a door closing state;
and if the line segment spacing of the line features after the weight removal is greater than or equal to the spacing threshold value, the rolling shutter door state is a door opening state.
6. The method of claim 1, wherein the de-duplicating the line characteristics after the line reconstruction comprises:
if the starting point and the ending point of the current line feature are the same as those of the previous line feature, deleting the current line feature, and reserving the previous line feature.
7. The rolling door state detection method according to claim 1, further comprising:
setting an image detection area, a length threshold value, a gradient threshold value, an angle threshold value, a distance threshold value and a line number threshold value of an LSD algorithm.
8. The rolling door state detection method according to claim 1, further comprising:
initializing a memory space, a length threshold, a gradient threshold, an angle threshold, a pitch threshold and a line number threshold of an LSD algorithm.
9. A rolling door condition detection system, comprising:
the acquisition module is used for acquiring the roller shutter door image;
the initialization module is used for carrying out projection transformation on the rolling shutter door image by adopting a predetermined transformation matrix;
the processing module is used for carrying out LSD algorithm processing on the image after projection transformation to obtain an algorithm detection result; calculating line characteristics according to the algorithm detection result; filtering the line feature; performing line reconstruction on the filtered line characteristics; performing de-duplication on the line characteristics after line reconstruction;
the result acquisition module is used for judging the state of the roller shutter door based on the line characteristics after the weight removal;
wherein the filtering the line feature comprises:
when the length of the roller shutter door image is not 0, deleting the line feature if the length of the line feature is calculated to be smaller than a length threshold value; otherwise, preserving the line feature;
when the gradient of the rolling door image is not 0, deleting the line feature if the gradient of the line feature is calculated to be smaller than a gradient threshold value; otherwise, preserving the line feature;
when the angle of the rolling door image is not 0, deleting the line feature if the angle of the line feature is calculated to be smaller than an angle threshold value; otherwise, preserving the line feature;
the line reconstruction of the filtered line features comprises the following steps:
changing an x coordinate corresponding to a y coordinate of the line feature starting point to 0, and changing an x coordinate corresponding to a y coordinate of the end point to an image detection area width RoI_w, wherein the formula is as follows:
original line characteristics:
f＝{(x s ,y s )，(x e ,y e )，len,ang,grad}
reconstruction post-line features:
f＝{(0,y s )，(RoI_w,y e )，len,ang,grad}
wherein x is s An x coordinate representing the line feature start point coordinate; y is s A y coordinate representing a line feature start coordinate; x is x e An x coordinate representing a line feature endpoint coordinate; y is e A y coordinate representing a line feature end point coordinate; roI_w represents the image detection area width; len represents the length of the line feature; ang represents the angle of the line feature; grad represents the gradient of the line feature.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011127173.1A CN112258471B (en) | 2020-10-20 | 2020-10-20 | Rolling door state detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011127173.1A CN112258471B (en) | 2020-10-20 | 2020-10-20 | Rolling door state detection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112258471A CN112258471A (en) | 2021-01-22 |
| CN112258471B true CN112258471B (en) | 2023-07-14 |
Family
ID=74244181
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011127173.1A Active CN112258471B (en) | 2020-10-20 | 2020-10-20 | Rolling door state detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112258471B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113516016B (en) * | 2021-04-21 | 2023-08-01 | 新疆爱华盈通信息技术有限公司 | Method and system for identifying opening and closing states of rolling gate of parking garage |
| CN113139458A (en) * | 2021-04-21 | 2021-07-20 | 新疆爱华盈通信息技术有限公司 | Method and system for identifying opening and closing states of parking garage roller shutter |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3640332A (en) * | 1969-11-12 | 1972-02-08 | Overhead Door Corp | Roll-type door |
| DE10320921A1 (en) * | 2003-05-09 | 2004-12-02 | Heinrich Strunz Gmbh | An emergency motorized door or window operating system has an outside control unit and sensors by which the door is opened or closed to a partially open condition |
| CN106986248A (en) * | 2017-05-23 | 2017-07-28 | 四川长虹电器股份有限公司 | Elevator switch door detection method based on photographed images |
| CN108569606A (en) * | 2018-06-15 | 2018-09-25 | 西安理工大学 | Construction elevator safety door angle identification method based on bounding box features |
| CN108674386A (en) * | 2018-07-10 | 2018-10-19 | 春笋新科技(深圳)有限公司 | A kind of Intelligent unattended car carwash, method and system |
| CN109035278A (en) * | 2018-07-25 | 2018-12-18 | 深圳市荣盛智能装备有限公司 | The detection method of fire exit door and its switch state based on image |
| CN209417582U (en) * | 2018-12-14 | 2019-09-20 | 无锡天奇信息技术有限公司 | A kind of control system of rolling screen door |
| CN111680621A (en) * | 2020-06-05 | 2020-09-18 | 浙江新再灵科技股份有限公司 | Elevator door state detection method based on video image |
| CN111723704A (en) * | 2020-06-09 | 2020-09-29 | 杭州古德微机器人有限公司 | Raspberry pie-based van body door opening monitoring method |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2300949B1 (en) * | 2008-05-22 | 2016-02-17 | Otis Elevator Company | Video-based system and method of elevator door detection |
| CN106285421A (en) * | 2016-10-25 | 2017-01-04 | 徐工消防安全装备有限公司 | Rolling screen door opening and closing detector, casing and fire fighting truck |
| CN109118687A (en) * | 2017-06-26 | 2019-01-01 | 霍尼韦尔国际公司 | The method for detecting possible threat/attack in ATM by extracting the deviation in ATM/roller shutter state |
| CN107449506A (en) * | 2017-08-30 | 2017-12-08 | 贵州省智能崛起科技有限公司 | A kind of vibration detection device and method for detecting vibration for rolling door axle |
| CN108584586A (en) * | 2017-11-21 | 2018-09-28 | 江苏鸿山鸿锦物联技术有限公司 | A kind of elevator door state line detection method based on video image |
| CN108805075B (en) * | 2018-06-07 | 2022-02-08 | 宁夏大学 | Driving trajectory line obtaining method and device and electronic equipment |
| CN109368434B (en) * | 2018-11-23 | 2020-09-01 | 张勇 | Elevator control system with video monitoring function |
| CN109978878A (en) * | 2019-04-04 | 2019-07-05 | 厦门商集网络科技有限责任公司 | Color image line segment detecting method and its system based on LSD |
| CN210768538U (en) * | 2019-07-24 | 2020-06-16 | 江苏盈丰电子科技有限公司 | Rolling slats door of measurable stroke speed and brake failure prevention |
-
2020
- 2020-10-20 CN CN202011127173.1A patent/CN112258471B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3640332A (en) * | 1969-11-12 | 1972-02-08 | Overhead Door Corp | Roll-type door |
| DE10320921A1 (en) * | 2003-05-09 | 2004-12-02 | Heinrich Strunz Gmbh | An emergency motorized door or window operating system has an outside control unit and sensors by which the door is opened or closed to a partially open condition |
| CN106986248A (en) * | 2017-05-23 | 2017-07-28 | 四川长虹电器股份有限公司 | Elevator switch door detection method based on photographed images |
| CN108569606A (en) * | 2018-06-15 | 2018-09-25 | 西安理工大学 | Construction elevator safety door angle identification method based on bounding box features |
| CN108674386A (en) * | 2018-07-10 | 2018-10-19 | 春笋新科技(深圳)有限公司 | A kind of Intelligent unattended car carwash, method and system |
| CN109035278A (en) * | 2018-07-25 | 2018-12-18 | 深圳市荣盛智能装备有限公司 | The detection method of fire exit door and its switch state based on image |
| CN209417582U (en) * | 2018-12-14 | 2019-09-20 | 无锡天奇信息技术有限公司 | A kind of control system of rolling screen door |
| CN111680621A (en) * | 2020-06-05 | 2020-09-18 | 浙江新再灵科技股份有限公司 | Elevator door state detection method based on video image |
| CN111723704A (en) * | 2020-06-09 | 2020-09-29 | 杭州古德微机器人有限公司 | Raspberry pie-based van body door opening monitoring method |
Non-Patent Citations (4)
| Title |
|---|
| Control model of the roller shutter on cloudy days;Xun Zhang et al;《2017 International Conference on Circuits,Devices and Systems》;84-88 * |
| From Bows to Arrows：Rolling Shutter Rectification of Urban Scenes;Vijay Rengarajan et al;《Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition》;2273-2781 * |
| 基于计算机视觉的电梯轿厢智能监控系统的研究与设计;金晓磊;《中国优秀硕士学位论文全文数据库（工程科技Ⅱ辑）》(第3期);C038-819 * |
| 基于计算机视觉的自主导航机器人门检测算法;陈祥;《计算机仿真》;第29卷(第6期);163-167 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112258471A (en) | 2021-01-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Conte et al. | A method for counting people in crowded scenes | |
| EP2118862B1 (en) | System and method for video detection of smoke and flame | |
| CN107256225B (en) | Method and device for generating heat map based on video analysis | |
| CN112258471B (en) | Rolling door state detection method and system | |
| CN109376631B (en) | Loop detection method and device based on neural network | |
| Debiasi et al. | PRNU variance analysis for morphed face image detection | |
| Hussain et al. | A comparative analysis of edge detection techniques used in flame image processing | |
| Batool et al. | Modeling and detection of wrinkles in aging human faces using marked point processes | |
| CN111898486B (en) | Monitoring picture abnormality detection method, device and storage medium | |
| Cheong et al. | A novel face detection algorithm using thermal imaging | |
| CN111445459A (en) | Image defect detection method and system based on depth twin network | |
| Ducottet et al. | Scale-adaptive detection and local characterization of edges based on wavelet transform | |
| US20210200990A1 (en) | Method for extracting image of face detection and device thereof | |
| CN108875504A (en) | Image detecting method and image detection device neural network based | |
| CN111444555B (en) | Temperature measurement information display method and device and terminal equipment | |
| CN115564710A (en) | Fire smoke detection method and device based on LK optical flow method and storage medium | |
| CN112258469A (en) | Rolling door state detection method and system | |
| Patil et al. | Chest X-ray features extraction for lung cancer classification | |
| Matsushita et al. | A probabilistic intensity similarity measure based on noise distributions | |
| CN113436735A (en) | Body weight index prediction method, device and storage medium based on face structure measurement | |
| CN111507347A (en) | Electrical equipment infrared image enhancement and segmentation method based on partial differential equation | |
| CN107153815A (en) | A kind of auth method, equipment and storage medium | |
| CN111401275B (en) | Information processing method and device for identifying grassland edge | |
| CN114140861A (en) | Face detection duplication removing method and device | |
| CN112183383A (en) | Information processing method and device for measuring face image in complicated environment |
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 |