CN114155262A - Somatosensory depth map acquisition system based on OPENCV correlation algorithm - Google Patents
Somatosensory depth map acquisition system based on OPENCV correlation algorithm Download PDFInfo
- Publication number
- CN114155262A CN114155262A CN202111436191.2A CN202111436191A CN114155262A CN 114155262 A CN114155262 A CN 114155262A CN 202111436191 A CN202111436191 A CN 202111436191A CN 114155262 A CN114155262 A CN 114155262A
- Authority
- CN
- China
- Prior art keywords
- module
- image
- data
- contour
- depth
- 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
- 230000003238 somatosensory effect Effects 0.000 title claims abstract description 35
- 238000005457 optimization Methods 0.000 claims abstract description 39
- 238000004364 calculation method Methods 0.000 claims abstract description 14
- 238000006243 chemical reaction Methods 0.000 claims description 42
- 238000001914 filtration Methods 0.000 claims description 27
- 238000013500 data storage Methods 0.000 claims description 20
- 238000013507 mapping Methods 0.000 claims description 13
- 239000003086 colorant Substances 0.000 claims description 5
- 239000002245 particle Substances 0.000 claims description 4
- 230000006870 function Effects 0.000 description 6
- 238000000034 method Methods 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000005286 illumination Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 2
- 150000002500 ions Chemical class 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229920006395 saturated elastomer Polymers 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- 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
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20132—Image cropping
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of OPEDCV, in particular to a motion sensing depth map acquisition system based on OPENCV related algorithm; the motion sensing device comprises a motion sensing device, an image optimization unit, a shape recognition unit, a color filling unit and a display; according to the invention, firstly, the somatosensory device interface is used for obtaining image depth data and sending the image depth data to the image optimization unit, then the image optimization unit is used for conducting depth cutting on the image data to obtain building block outlines in the image, the image outline data is sent to the shape recognition unit, the shape recognition unit conducts binarization, outline searching and polygon fitting on the image information to obtain the number of vertexes of each outline, area calculation is conducted on the outlines, redundant outlines are removed, and finally the information is sent to the display for displaying.
Description
Technical Field
The invention relates to the technical field of OPEDCV, in particular to a motion sensing depth map acquisition system based on OPENCV related algorithm.
Background
OpenCV is a BSD license (open source) based distributed cross-platform computer vision and machine learning software library that can run on Linux, Windows, Android, and MacOS operating systems. The method is light and efficient, is composed of a series of C functions and a small number of C + + classes, provides interfaces of languages such as Python, Ruby, MATLAB and the like, and realizes a plurality of general algorithms in the aspects of image processing and computer vision.
Most of the existing somatosensory depth map acquisition systems rely on RGB images, and the quality of the RGB images is easily influenced by factors such as strong light, severe illumination intensity change, multi-object overlapping shielding, data transmission loss and the like, so that an algorithm or a model is limited, the adaptability is low, most importantly, the motion of people is complex, and the identification precision cannot meet the requirement due to the fact that some methods are too simple.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a motion sensing depth map acquisition system based on an OPENCV related algorithm, firstly, a motion sensing device interface is used for acquiring depth data and sending the depth data to an image optimization unit, the image optimization unit is used for performing depth clipping on the image data, then, the depth of a pixel of a depth clipping layer is subjected to linear mapping to acquire lineDepth parameters, the lineDepth parameters are adjusted to filter edge noise points, the image data is sent to a shape recognition unit after the noise points are removed, the shape recognition unit performs binarization-contour searching-polygon fitting on the image information to acquire the number of vertexes of each contour, performs data shape recognition on different contours according to the acquired number of fixed points, removes redundant contours, then sends the information to a display for displaying, and changes the depth data by clipping on the whole use, the method avoids the influence of factors such as strong light, severe illumination intensity change, multi-object overlapping shielding, data transmission loss and the like, and further improves the applicability of the algorithm and the model.
In order to achieve the purpose, the invention provides the following technical scheme:
a somatosensory depth map acquisition system based on an OPENCV correlation algorithm comprises:
a body sensing device to obtain depth data;
the image optimization unit is used for receiving data information sent by the somatosensory equipment and acquiring a somatosensory depth map;
the shape recognition unit is used for receiving the image optimization unit and acquiring image contour information from data in the somatosensory depth map;
the color filling unit is used for receiving the information sent by the shape recognition unit and filling colors for the image outline information;
and the display is used for receiving and displaying the image information sent by the shape recognition unit and the color filling unit.
The invention is further configured to: body sensing equipment and image optimization unit electric connection, image optimization unit and shape recognition unit electric connection, shape recognition unit and color fill unit all with display electric connection.
The invention is further configured to: the image optimization unit includes a depth recognition module and a depth information storage module, wherein:
the depth recognition module is used for acquiring a depth interval of the image and is electrically connected with the somatosensory equipment;
the depth information storage module is used for receiving and storing the information sent by the depth identification module, and the depth information storage module is electrically connected with the depth identification module.
The invention is further configured to: the image optimization unit further comprises a linear mapping module and a noise filtering module, wherein:
the linear mapping module is used for acquiring linear depths of a plurality of spatial planes;
and the noise filtering module is used for filtering redundant linear particles outside the linear mapping module.
The invention is further configured to: the shape recognition unit comprises a pixel conversion module, a data storage module and a data search module, wherein:
the pixel conversion module is used for receiving the image information sent by the image optimization unit and converting an image original image into pixel data for gray value conversion, and is electrically connected with the image optimization unit;
the data storage module is used for receiving gray value data information generated by the data storage module, and the data storage module is electrically connected with the pixel conversion module;
the data searching module is used for acquiring outline pixel data information of the communicated region and is electrically connected with the data storage module.
The invention is further configured to: the shape recognition unit further comprises a fitting module, a contour information storage module and a contour filtering module, wherein:
the fitting module is used for obtaining contour convex point information, calculating the contour shape of the image and sending the convex point information and the image contour information to the contour information storage module;
the contour information storage module is used for receiving and storing the data information sent by the fitting module, and is electrically connected with the fitting module;
the contour filtering module is used for calculating the contour area stored in the contour information storage module and filtering out a larger contour and a smaller contour.
The invention is further configured to: the color filling unit comprises a color information storage module, a central point calculation module and a pixel color conversion module, wherein:
the color information storage module is used for storing the image information after color change and is electrically connected with the shape recognition unit;
the central point calculation module is used for calculating central point data of the minimum circumcircle of the outline and sending the central point data to the pixel color conversion module;
and the pixel color conversion module is used for receiving the data information sent by the central point calculation module and performing color conversion.
The invention is further configured to: the color filling unit further comprises an image color conversion module and a distinguishing module, wherein:
the image color conversion module is used for changing the integral pixel color of the image outline;
the distinguishing module is used for changing the tone of the contour image so as to distinguish a plurality of image contours.
Advantageous effects
Compared with the known public technology, the technical scheme provided by the invention has the following beneficial effects:
the invention firstly uses a somatosensory device interface to take depth data with the size of 512x424 and send the data to an image optimization unit, then uses the image optimization unit to carry out depth clipping on the image data to obtain a building block outline in an image for adjusting the depth clipping range, then carries out linear mapping on the pixel depth of a depth clipping layer to obtain lineDepth parameters, adjusts the lineDepth parameters to filter edge noise points, sends the image data to a shape recognition unit after the noise points are removed, the shape recognition unit carries out binarization, contour searching and polygon fitting on the image information to obtain the vertex number of each outline, carries out data shape recognition on different outlines according to the obtained fixed point number, carries out area calculation on the outlines, removes redundant outlines, improves the using effect of the image, then sends the information to a display for showing, is integrally used, and changes the depth data by clipping, the method avoids the influence of factors such as strong light, severe illumination intensity change, multi-object overlapping shielding, data transmission loss and the like, and further improves the applicability of the algorithm and the model.
Drawings
FIG. 1 is a system diagram of a somatosensory depth map acquisition system based on an OPENCV correlation algorithm;
FIG. 2 is a system diagram of the interior of an image optimization unit in a somatosensory depth map acquisition system based on an OPENCV correlation algorithm;
FIG. 3 is a system diagram of the interior of a shape recognition unit in a somatosensory depth map acquisition system based on an OPENCV correlation algorithm;
fig. 4 is a system diagram of the interior of a color filling unit in a somatosensory depth map acquisition system based on an OPENCV correlation algorithm.
In the figure: 1. a motion sensing device; 2. an image optimization unit; 201. a depth recognition module; 202. A depth information storage module; 203. a linear mapping module; 204. a noise filtering module; 3. A shape recognition unit; 301. a pixel conversion module; 302. a data storage module; 303. a data search module; 304. a fitting module; 305. a contour information storage module; 306. a contour filtering module; 4. a color filling unit; 401. a color information storage module; 402. a central point calculation module; 403. a pixel color conversion module; 404. an image color conversion module; 405. A distinguishing module; 5. a display.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The present invention will be further described with reference to the following examples.
Example 1
Referring to fig. 1-4, a somatosensory depth map acquisition system based on OPENCV correlation algorithm includes:
the motion sensing device 1 is used for acquiring depth data;
the image optimization unit 2 is used for receiving data information sent by the somatosensory device 1 and acquiring a somatosensory depth map;
the shape recognition unit 3 is used for receiving the image optimization unit 2 and acquiring image contour information from data in the somatosensory depth map;
the color filling unit 4 is used for receiving the information sent by the shape recognition unit 3 and filling colors for the image outline information;
the display 5, the display 5 is used for receiving the image information sent by shape recognition unit 3 and color filling unit 4 and displaying;
When the invention is used, firstly depth data (which can directly obtain data through a somatosensory native interface) with the size of 512x424 is taken by using an interface of a somatosensory device 1 and is sent to an image optimization unit 2, then the image optimization unit 2 is used for carrying out depth clipping on the image data to obtain a building block outline in an image, two parameters of minDepth and maxDepth are used for adjusting a depth clipping range, then the pixel depth of a depth clipping layer is linearly mapped to obtain a lineDepth parameter, the lineDepth parameter is adjusted to filter edge noise points, the image data is sent to a shape recognition unit 3 after the noise points are removed, the shape recognition unit 3 carries out binarization-contour searching-polygon fitting on the image information to obtain the number of vertexes of each outline, carries out data shape recognition on different outlines according to the obtained number of fixed points, carries out area calculation on the outlines and removes redundant outlines, improve the image result of use, later with information transmission to color filling unit 4, send to display 5 simultaneously and demonstrate, when information reachd color filling unit 4, color filling unit 4 carries out the central point calculation to each profile, and carry out color conversion with the central point, later use the central point as the center and carry out color conversion to the profile, carry out tone adjustment to the profile map at last, send display 5 after the regulation is accomplished and demonstrate, in the whole use, change through tailorring the degree of depth data, avoided receiving the highlight, violent illumination intensity changes, the influence of factors such as many objects overlap shelters from and data transmission loss, and then the suitability of algorithm and model has been improved.
The image optimization unit 2 comprises a depth recognition module 201 and a depth information storage module 202, wherein:
the depth recognition module 201 is used for acquiring a depth interval of an image, and the depth recognition module 201 is electrically connected with the somatosensory device 1;
the depth information storage module 202 is configured to receive and store information sent by the depth identification module 201, and the depth information storage module 202 is electrically connected to the depth identification module 201.
The image optimization unit 2 further comprises a linear mapping module 203 and a noise filtering module 204, wherein:
the linear mapping module 203 is used for acquiring linear depths of a plurality of spatial planes;
the noise filtering module 204 is used for filtering the redundant linear particles outside the linear mapping module 203.
In the present invention, when the depth data map reaches the image optimization unit 2, the depth recognition module 201 is used to obtain a single data depth value range [0-5000], and the larger the value is, the farther the distance from the camera is, the depth range rule is: the depth value of the building block is in a certain range, such as [3500-, by filtering the pixel ion data of different planes in the image, the outline of the image information generated after filtering is clearer.
The shape recognition unit 3 includes a pixel conversion module 301, a data storage module 302, and a data lookup module 303, wherein:
the pixel conversion module 301 is configured to receive the image information sent by the image optimization unit 2, convert an image original image into pixel data, and perform gray value conversion, where the pixel conversion module 301 is electrically connected to the image optimization unit 2;
the data storage module 302 is configured to receive gray-level data information generated by the data storage module 302, and the data storage module 302 is electrically connected to the pixel conversion module 301;
the data search module 303 is configured to obtain the contour pixel data information of the connected region, and the data search module 303 is electrically connected to the data storage module 302.
The shape recognition unit 3 further comprises a fitting module 304, a contour information storage module 305 and a contour filtering module 306, wherein:
the fitting module 304 is configured to obtain the contour convex point information, calculate the contour shape of the image, and send the convex point information and the contour information of the image to the contour information storage module 305;
the contour information storage module 305 is configured to receive and store the data information sent by the fitting module 304, and the contour information storage module 305 is electrically connected to the fitting module 304;
the contour filtering module 306 is used to calculate the contour area stored in the contour information storage module 305 and filter out larger contours and smaller contours.
In the invention, when the image optimization unit 2 sends information to the shape recognition unit 3, firstly, the pixel conversion module 301 is used to convert the pixel information, the gray value is given, the converted image information is sent to the data storage module 302 to be stored, then the data search module 303 is used to count the image contour, the pixel data information in the data storage module 302 is traversed line by line to obtain the contour point of the connected region, the model contour is obtained, the data in the data storage module 302 is updated and replaced, the fitting module 304 is used to obtain the contour salient point by using the appaxpolydp, the salient point data is collected, the minimum bounding rectangle of the contour is calculated by using the miniAreact, the aspect ratio is calculated to obtain whether the rectangle is square (the height is more than 1.25 is the rectangle) (the number of the salient points is more than 5 is the prototype), the redundant pixels are not passed through when the shape of the salient point is fitted, and sends the information to the profile information storage module 305, and finally, the profile filtering module 306 calculates different profile areas in the profile information storage module 305 by using a contourArea function, filters out profile information with overlarge or undersize profile areas, and in the aspect of overall use, calculates and filters different profile data of the information of the image, thereby improving the accuracy of the image information.
The color filling unit 4 includes a color information storage module 401, a center point calculation module 402, and a pixel color conversion module 403, wherein:
the color information storage module 401 is used for storing the image information after color change, and the color information storage module 401 is electrically connected with the shape recognition unit 3;
the central point calculating module 402 is configured to calculate central point data of the minimum circumcircle of the contour and send the central point data to the pixel color converting module 403;
the pixel color conversion module 403 is configured to receive the data information sent by the central point calculation module 402 and perform color conversion.
The color filling unit 4 further comprises an image color conversion module 404 and a differentiating module 405, wherein:
the image color conversion module 404 is used to change the whole pixel color of the image contour;
the distinguishing module 405 is used to change the color bands of the contour image to facilitate distinguishing a plurality of image contours.
When the shape recognition unit 3 sends information to the color filling unit 4, firstly, the information is stored through the color information storage module 401, then, the central point calculation module 402 uses minEnclosingCircle to calculate the central point of the minimum circumcircle of the outline and stores the information into the color information storage module 401, the pixel color conversion module 403 uses the somatosensory interface MapDepthPointToColorCoords to convert the central point coordinate into a color coordinate (a depth image is converted into a color image) and store the color image into the color information storage module 401, then, the image color conversion module 404 uses cvtColor to convert the somatosensory color from an RGB space into H (Hue) S (saturated) V (lightness) space and cover the stored information in the color information storage module 401, and finally, the distinguishing module 405 uses a mean function to calculate the average pixel Hue value of a kernel with the size of 3x3 and taking the central point of the outline as the center, the colors are distinguished according to Hue values, and then the building blocks are distinguished, and in the aspect of overall use, a user can select whether to execute the color filling unit 4 to perform color filling work according to the use environment so as to meet the requirements of various real environments.
When the invention works, firstly depth data (which can be directly obtained through a somatosensory native interface) with the size of 512x424 is taken by using an interface of a somatosensory device 1 and is sent to an image optimization unit 2, then a depth recognition module 201 is used, the obtained single data depth value range is [0-5000], only pixel values of the data values in the range [ minDepth, maxDepth ] are reserved, the reserved pixel values are sent to a depth information storage module 202 for storage, then a linear mapping module 203 is used for combining linear particles with the same pixel information in a plurality of parallel planes into a linear matrix and sending the linear matrix to the depth information storage module 202 for storage, finally a noise point filtering module 204 is used for obtaining linear depth information according to the edge of a graph, unexpected linear ions outside the contour of closed information are filtered according to the linear depth, and accordingly the contour display of multiple planes is improved, reducing the interference of perspective, after the image optimization unit 2 finishes working, the noise point filtering module 204 will send information to the shape recognition unit 3, convert the pixel information by using the pixel conversion module 301, give a gray value and send the converted image information to the data storage module 302 for storage, then count the image contour by using findContours through the data search module 303, then obtain the contour convex point by using the propaxploydp through the fitting module 304, collect the convex point data, calculate the contour maximum by using minAreaRect, do not pass through the redundant pixels when fitting according to the shape of the convex point image, send the information to the contour information storage module 305, finally calculate different contour areas in the contour information storage module 305 by using the contourea function through the contour filtering module 306, filter out the contour information if the contour area is too large or too small, send the information to the display 5 by using the contour filtering module 306 after filtering is finished, by means of the display 5, it is shown that, when the color filling cell 4 needs to be used, information is sent to the shape recognition unit 3 to send information to the color filling cell 4, the color information is stored by a color information storage module 401, then the center point of the minimum circumcircle of the outline is calculated by a center point calculation module 402 by using minEnclosingCircle and the information is stored in the color information storage module 401, the motion sensing interface mapdepthpointtocolorcords is used by the pixel color conversion module 403 to convert the center point coordinates into color image coordinates, and storing the color data to a color information storage module 401, converting the somatosensory color from an RGB space to an HSV space by using a cvtColor through an image color conversion module 404, covering the stored information in the color information storage module 401, and finally distinguishing the colors according to Hue values by using a mean function through a distinguishing module 405 so as to distinguish the building blocks.
Portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof, and in the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system, for example, if implemented in hardware, and in another embodiment, any one or a combination of the following techniques, as is known in the art: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit and scope of the corresponding technical solutions.
Claims (8)
1. A motion sensing depth map acquisition system based on an OPENCV correlation algorithm is characterized by comprising:
the motion sensing device (1), wherein the motion sensing device (1) is used for acquiring depth data;
the image optimization unit (2) is used for receiving data information sent by the somatosensory device (1) and acquiring the somatosensory depth map;
the shape recognition unit (3) is used for receiving the image optimization unit (2) and acquiring image contour information from data in the somatosensory depth map;
the color filling unit (4), the color filling unit (4) is used for receiving the information sent by the shape recognition unit (3) and filling colors for the image outline information;
and the display (5), the display (5) is used for receiving and displaying the image information sent by the shape recognition unit (3) and the color filling unit (4).
2. The system for acquiring the motion sensing depth map based on the OPENCV related algorithm according to claim 1, wherein the motion sensing device (1) is electrically connected with an image optimization unit (2), the image optimization unit (2) is electrically connected with a shape recognition unit (3), the shape recognition unit (3) is electrically connected with a color filling unit (4), and the shape recognition unit (3) and the color filling unit (4) are both electrically connected with a display (5).
3. The somatosensory depth map acquisition system based on OPENCV correlation algorithm according to claim 1, wherein the image optimization unit (2) comprises a depth recognition module (201) and a depth information storage module (202), wherein:
the depth recognition module (201) is used for acquiring a depth interval of an image, and the depth recognition module (201) is electrically connected with the somatosensory device (1);
the depth information storage module (202) is used for receiving and storing information sent by the depth identification module (201), and the depth information storage module (202) is electrically connected with the depth identification module (201).
4. The somatosensory depth map acquisition system based on an OPENCV correlation algorithm according to claim 3, wherein the image optimization unit (2) further comprises a linear mapping module (203) and a noise filtering module (204), wherein:
the linear mapping module (203) is used for acquiring linear depths of a plurality of spatial planes;
the noise filtering module (204) is used for filtering redundant linear particles outside the linear mapping module (203).
5. The somatosensory depth map acquisition system based on an OPENCV correlation algorithm as claimed in claim 1, wherein the shape recognition unit (3) comprises a pixel conversion module (301), a data storage module (302) and a data lookup module (303), wherein:
the pixel conversion module (301) is used for receiving the image information sent by the image optimization unit (2) and converting an image original image into pixel data for gray value conversion, and the pixel conversion module (301) is electrically connected with the image optimization unit (2);
the data storage module (302) is used for receiving the gray value data information generated by the data storage module (302), and the data storage module (302) is electrically connected with the pixel conversion module (301);
the data searching module (303) is used for acquiring contour pixel data information of the connected region, and the data searching module (303) is electrically connected with the data storage module (302).
6. The somatosensory depth map acquisition system based on OPENCV correlation algorithm as claimed in claim 5, wherein the shape recognition unit (3) further comprises a fitting module (304), a contour information storage module (305) and a contour filtering module (306), wherein:
the fitting module (304) is used for obtaining contour convex point information, calculating the contour shape of the image and sending the convex point information and the contour information of the image to the contour information storage module (305);
the contour information storage module (305) is used for receiving and storing the data information sent by the fitting module (304), and the contour information storage module (305) is electrically connected with the fitting module (304);
the contour filtering module (306) is used for calculating the contour area stored in the contour information storage module (305) and filtering out a larger contour and a smaller contour.
7. The somatosensory depth map acquisition system based on OPENCV correlation algorithm according to claim 1, wherein the color filling unit (4) comprises a color information storage module (401), a central point calculation module (402) and a pixel color conversion module (403), wherein:
the color information storage module (401) is used for storing the image information after color change, and the color information storage module (401) is electrically connected with the shape recognition unit (3);
the central point calculating module (402) is used for calculating central point data of the minimum circumcircle of the outline and sending the central point data to the pixel color conversion module (403);
the pixel color conversion module (403) is used for receiving the data information sent by the central point calculation module (402) and performing color conversion.
8. The motion sensing depth map acquisition system based on OPENCV correlation algorithm as claimed in claim 7, wherein the color filling unit (4) further comprises an image color conversion module (404) and a distinguishing module (405), wherein:
the image color conversion module (404) is used for changing the integral pixel color of the image outline;
the distinguishing module (405) is used for changing the tone of the contour image so as to distinguish a plurality of image contours.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111436191.2A CN114155262A (en) | 2021-11-29 | 2021-11-29 | Somatosensory depth map acquisition system based on OPENCV correlation algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111436191.2A CN114155262A (en) | 2021-11-29 | 2021-11-29 | Somatosensory depth map acquisition system based on OPENCV correlation algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114155262A true CN114155262A (en) | 2022-03-08 |
Family
ID=80784386
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111436191.2A Pending CN114155262A (en) | 2021-11-29 | 2021-11-29 | Somatosensory depth map acquisition system based on OPENCV correlation algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114155262A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104835164A (en) * | 2015-05-11 | 2015-08-12 | 京东方科技集团股份有限公司 | Method and device for processing binocular camera depth image |
| CN107750370A (en) * | 2015-06-16 | 2018-03-02 | 皇家飞利浦有限公司 | For the method and apparatus for the depth map for determining image |
| CN110378944A (en) * | 2019-07-11 | 2019-10-25 | Oppo广东移动通信有限公司 | Depth map processing method, device and electronic equipment |
| US20210207952A1 (en) * | 2018-09-28 | 2021-07-08 | Panasonic Intellectual Property Management Co., Ltd. | Measurement device and measurement method |
-
2021
- 2021-11-29 CN CN202111436191.2A patent/CN114155262A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104835164A (en) * | 2015-05-11 | 2015-08-12 | 京东方科技集团股份有限公司 | Method and device for processing binocular camera depth image |
| CN107750370A (en) * | 2015-06-16 | 2018-03-02 | 皇家飞利浦有限公司 | For the method and apparatus for the depth map for determining image |
| US20180295340A1 (en) * | 2015-06-16 | 2018-10-11 | Koninklijke Philips N.V. | Method and apparatus for determining a depth map for an angle |
| US20210207952A1 (en) * | 2018-09-28 | 2021-07-08 | Panasonic Intellectual Property Management Co., Ltd. | Measurement device and measurement method |
| CN110378944A (en) * | 2019-07-11 | 2019-10-25 | Oppo广东移动通信有限公司 | Depth map processing method, device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Corke et al. | Dealing with shadows: Capturing intrinsic scene appearance for image-based outdoor localisation | |
| US20220189017A1 (en) | Medical image processing method and apparatus, image processing method and apparatus, terminal and storage medium | |
| GB2543749A (en) | 3D scene rendering | |
| CN113192179B (en) | Three-dimensional reconstruction method based on binocular stereo vision | |
| CN114022830A (en) | Target determination method and target determination device | |
| CN103426200B (en) | Tree three-dimensional reconstruction method based on unmanned aerial vehicle aerial photo sequence image | |
| KR20050022306A (en) | Method and Apparatus for image-based photorealistic 3D face modeling | |
| CN116309607B (en) | Ship type intelligent water rescue platform based on machine vision | |
| CN106504289B (en) | indoor target detection method and device | |
| US11282180B1 (en) | Object detection with position, pose, and shape estimation | |
| CN117079117B (en) | Underwater image processing and target identification method and device, storage medium and electronic equipment | |
| CN113570701B (en) | Hair reconstruction method and device | |
| CN114639115A (en) | 3D pedestrian detection method based on fusion of human body key points and laser radar | |
| CN108154496A (en) | A kind of power equipment appearance suitable for electric operating robot changes recognition methods | |
| EP3144893A1 (en) | Method and apparatus for estimating reflectance parameters and a position of a light source in a scene | |
| CN112598674B (en) | Image processing method and device for vehicle and vehicle | |
| JP2020139877A (en) | Information processing device, information processing method and program | |
| Petrovai et al. | Obstacle detection using stereovision for Android-based mobile devices | |
| CN110514140B (en) | Three-dimensional imaging method, device, equipment and storage medium | |
| CN114155262A (en) | Somatosensory depth map acquisition system based on OPENCV correlation algorithm | |
| CN113077523A (en) | Calibration method, calibration device, computer equipment and storage medium | |
| CN112819935A (en) | Method for realizing three-dimensional reconstruction of workpiece based on binocular stereo vision | |
| CN117292076A (en) | Dynamic three-dimensional reconstruction method and system for local operation scene of engineering machinery | |
| WO2021118386A1 (en) | Method for obtaining a set of objects of a three-dimensional scene | |
| CN113808262B (en) | Building model generation system based on depth map analysis |
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 |