CN111524151A - Object detection method and system based on background recognition - Google Patents
Object detection method and system based on background recognition Download PDFInfo
- Publication number
- CN111524151A CN111524151A CN202010357352.8A CN202010357352A CN111524151A CN 111524151 A CN111524151 A CN 111524151A CN 202010357352 A CN202010357352 A CN 202010357352A CN 111524151 A CN111524151 A CN 111524151A
- Authority
- CN
- China
- Prior art keywords
- outline
- contour
- background
- background environment
- inner frame
- 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
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
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V8/00—Prospecting or detecting by optical means
- G01V8/10—Detecting, e.g. by using light barriers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/181—Segmentation; Edge detection involving edge growing; involving edge linking
-
- 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
- G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- 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
- G06T2207/10004—Still image; Photographic image
-
- 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
- G06T2207/20036—Morphological image processing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Geometry (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geophysics (AREA)
- Image Analysis (AREA)
Abstract
The invention provides an object detection method based on background recognition, which comprises the following steps: shooting an image in a set background environment, wherein the background environment comprises an inner frame outline and an outer frame outline which has the same central point as the inner frame outline, and an object to be detected is preset in the outer frame outline and covers the inner frame outline; and extracting and identifying the contour in the image, and if the inner frame contour is not detected and the outer frame contour is detected, judging that an object exists in the background environment. The object detection method based on background recognition has the advantage of accurate detection.
Description
Technical Field
The invention relates to the field of object detection, in particular to an object detection method and system based on background recognition.
Background
In some application scenarios, such as applications of warehousing and express delivery, it is necessary to detect the placement position of an object to determine whether the object is placed at a specific position or whether the object itself has a problem. As methods for detecting an object, object detection by an ultrasonic sensor and object detection by a laser sensor have been common.
The principle of object detection based on ultrasonic sensors is as follows: the ultrasonic sensor is composed of an ultrasonic transmitter and an ultrasonic receiver, and is integrated with the transmitting and receiving. And starting ultrasonic transmission timing, and calculating the distance between the measured object and the sensor according to the sound propagation speed when the receiver detects that the echo reflected by the measured object finishes timing. The presence of the object is judged by the distance.
The principle disadvantage of object detection based on ultrasonic sensors is: when the surface of the measured object is rough or the detected surface is not over against the ultrasonic sensor, the receiver can not receive the sound wave returned by the measured object, and a certain detection blind area exists.
Principle of object detection based on laser sensor: the laser sensor consists of a laser emitting diode and a receiver. When the laser sensor works, the laser emitting diode is aligned to the detected object to emit laser pulses. After being reflected by the object, the laser is scattered to all directions, and part of scattered light returns to the sensor receiver. The method judges whether an object exists by judging whether a sensor receiver receives scattered light.
Disadvantages of object detection based on laser sensors: when the object is an irregular object, the scattered light does not necessarily return to the sensor receiver, which will not ensure the accuracy of object detection. And the laser sensor is greatly disturbed by the color of the detected object.
Disclosure of Invention
The invention aims to provide an object detection method and system based on background recognition, which are accurate in detection.
In an embodiment of the present invention, an object detection method based on background recognition is provided, which includes:
shooting an image in a set background environment, wherein the background environment comprises an inner frame outline and an outer frame outline which has the same central point as the inner frame outline, and an object to be detected is preset in the outer frame outline and covers the inner frame outline;
and extracting and identifying the contour in the image, and if the inner frame contour is not detected and the outer frame contour is detected, judging that an object exists in the background environment.
In the embodiment of the invention, if the inner frame outline and the outer frame outline are detected, it is judged that no object exists in the background environment.
In the embodiment of the invention, if the outline of the outer frame is not detected, the background environment is judged to be abnormal.
In the embodiment of the invention, the inner frame outline and the outer frame outline are both rectangular or circular.
In the embodiment of the present invention, extracting a contour in an image includes:
converting the shot image into a gray image;
extracting the gray level image for edge detection;
communicating the interior of the edge obtained by the edge detection and carrying out corrosion treatment;
and extracting the contour and contour characteristic data of the obtained image.
In the embodiment of the present invention, for an object with a simple appearance shape, after determining that an object exists in the background environment, the method further includes:
extracting the contour of the object, comparing the contour of the object with the actual contour of the object sample,
if the size of the outline of the object is not consistent with the size of the actual outline of the object sample, judging that the object has abnormal appearance shape and giving an alarm for prompting;
and if the outline of the object is inconsistent with the preset position of the actual outline of the object sample, judging that the placement position of the object deviates from the preset position and giving an alarm for prompting.
In an embodiment of the present invention, the method for detecting an object based on background recognition further includes:
and calibrating the background environment before using the background environment.
In the embodiment of the present invention, calibrating the background environment includes:
and under the condition that no object exists, a fixed camera is arranged to acquire and process the image of the background environment, the inner frame outline and the outer frame outline are extracted, whether the characteristic parameters of the inner frame outline and the outer frame outline meet preset values or not is judged, if yes, the camera and the background environment meet the requirements, and if not, the position of the camera or the background environment needs to be adjusted.
The embodiment of the invention also provides an object detection system based on background recognition, which adopts the object detection method based on background recognition to detect the object.
Compared with the prior art, the object detection method based on background recognition is adopted to shoot an image in a set background environment, the background environment comprises an inner frame outline and an outer frame outline which has the same central point as the inner frame outline, an object to be detected is placed in the outer frame outline in a preset mode and covers the inner frame outline, then the outline in the image is extracted and recognized, if the inner frame outline is not detected and the outer frame outline is detected, the object in the background environment is judged to exist, whether the object exists in the set background environment can be accurately recognized, and the realization process is simple.
Drawings
FIG. 1 is a schematic illustration of a background environment for object detection in an embodiment of the present invention.
Fig. 2 is a flowchart of an object detection method based on background recognition according to an embodiment of the present invention.
FIG. 3 is a flow chart of extracting a contour according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The following describes the implementation of the present invention in detail with reference to specific embodiments.
The embodiment of the invention provides an object detection method based on background identification, wherein an object to be detected needs to be identified in a set background environment.
As shown in fig. 1, the background environment includes an inner frame contour and an outer frame contour having the same center point as the inner frame contour, and the object to be detected is preset in the outer frame contour and covers the inner frame contour. The outline area of the outer frame is a limited area for placing objects, and the area can not be covered or shielded in actual use. The inner frame outline area is an object detection area which must be shielded when an object is placed. The inner frame contour and the outer frame contour may be rectangular, circular or other shapes, and the shapes of the inner frame contour and the outer frame contour may be the same or different, which is not limited in the present invention.
It should be noted that the object to be detected is an object specified in a certain application scenario, has a uniform shape characteristic, and also needs to be placed at a specified position, usually at a central position in the background environment, and the outline of the object to be detected is larger than the outline of the inner frame and smaller than the outline of the outer frame, so as to ensure that the object to be detected can be accurately identified. For example, in a round turntable type storage mechanism, lunch boxes are regularly placed on a turntable, whether the lunch boxes exist on the turntable needs to be detected, whether the lunch boxes exist in the background environment can be identified by arranging the background environment on the turntable through a camera, and if the inner frame contour is not detected and the outer frame contour is detected, it is determined that objects exist in the background environment.
As shown in fig. 2, the background environment is calibrated before being used. Calibrating the background environment comprises: and under the condition that no object exists, a fixed camera is arranged to acquire and process the image of the background environment, the inner frame outline and the outer frame outline are extracted, whether the characteristic parameters of the inner frame outline and the outer frame outline meet preset values or not is judged, if yes, the camera and the background environment meet the requirements, and if not, the position of the camera or the background environment needs to be adjusted. Specifically, if the outline of the outer frame cannot be found, the background environment is abnormal or the camera is abnormal; if the sizes of the inner frame outline and the outer frame outline do not accord with the preset value, the camera mounting distance is abnormal, and the camera needs to be mounted again; and if the relative position of the inner frame outline and the outer frame outline does not accord with a preset value, indicating that the installation angle of the camera is abnormal and needing to be corrected.
After the background environment calibration is completed, the method can be used for detecting the object, and the detection process is as follows:
shooting an image in the background environment, extracting a contour in the image and identifying;
if the inner frame contour is not detected and the outer frame contour is detected, judging that an object exists in the background environment; if the inner frame contour and the outer frame contour are detected, judging that no object exists in the background environment; and if the outline of the outer frame is not detected, judging that the background environment is abnormal.
For an object with a simple appearance shape, after judging that the object exists in the background environment, continuing to detect as follows:
extracting the contour of the object, comparing the contour of the object with the actual contour of the object sample,
if the size of the outline of the object is not consistent with the size of the actual outline of the object sample, judging that the object has abnormal appearance shape and giving an alarm for prompting;
and if the outline of the object is inconsistent with the preset position of the actual outline of the object sample, judging that the placement position of the object deviates from the preset position and giving an alarm for prompting.
As shown in fig. 3, extracting the contour in the image specifically includes:
converting the shot image into a gray image;
extracting the gray level image for edge detection;
communicating the interior of the edge obtained by the edge detection and carrying out corrosion treatment;
and extracting the contour and contour characteristic data of the obtained image.
The embodiment of the invention also provides an object detection system based on background recognition, which adopts the object detection method based on background recognition to detect the object.
In summary, according to the object detection method based on background recognition of the present invention, an image is captured in a set background environment, the background environment includes an inner frame contour and an outer frame contour having a same central point as the inner frame contour, an object to be detected is placed in the outer frame contour in a predetermined manner and covers the inner frame contour, then a contour in the image is extracted and recognized, if the inner frame contour is not detected and the outer frame contour is detected, it is determined that an object exists in the background environment, it is possible to accurately recognize whether an object exists in the set background environment, and the implementation process is simple.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (9)
1. An object detection method based on background recognition is characterized by comprising the following steps:
shooting an image in a set background environment, wherein the background environment comprises an inner frame outline and an outer frame outline which has the same central point as the inner frame outline, and an object to be detected is preset in the outer frame outline and covers the inner frame outline;
and extracting and identifying the contour in the image, and if the inner frame contour is not detected and the outer frame contour is detected, judging that an object exists in the background environment.
2. The object detection method based on background recognition according to claim 1, wherein if both the inner frame contour and the outer frame contour are detected, it is determined that no object exists in the background environment.
3. The object detection method based on background recognition of claim 1, wherein if the outline of the outer frame is not detected, it is determined that the background environment is abnormal.
4. The object detection method based on background recognition according to claim 1, wherein the inner frame profile and the outer frame profile are both rectangular or circular.
5. The background recognition-based object detection method of claim 1, wherein extracting the contour in the image comprises:
converting the shot image into a gray image;
extracting the gray level image for edge detection;
communicating the interior of the edge obtained by the edge detection and carrying out corrosion treatment;
and extracting the contour and contour characteristic data of the obtained image.
6. The object detection method based on background recognition according to claim 1, wherein, for an object with a simple appearance shape, after determining that an object exists in the background environment, the method further comprises:
extracting the contour of the object, comparing the contour of the object with the actual contour of the object sample,
if the size of the outline of the object is not consistent with the size of the actual outline of the object sample, judging that the object has abnormal appearance shape and giving an alarm for prompting;
and if the outline of the object is inconsistent with the preset position of the actual outline of the object sample, judging that the placement position of the object deviates from the preset position and giving an alarm for prompting.
7. The background recognition-based object detection method of claim 1, further comprising:
and calibrating the background environment before using the background environment.
8. The background recognition-based object detection method of claim 7, wherein calibrating the background environment comprises:
and under the condition that no object exists, a fixed camera is arranged to acquire and process the image of the background environment, the inner frame outline and the outer frame outline are extracted, whether the characteristic parameters of the inner frame outline and the outer frame outline meet preset values or not is judged, if yes, the camera and the background environment meet the requirements, and if not, the position of the camera or the background environment needs to be adjusted.
9. An object detection system based on background recognition, characterized in that the object is detected by the object detection method based on background recognition according to any one of claims 1-9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010357352.8A CN111524151A (en) | 2020-04-29 | 2020-04-29 | Object detection method and system based on background recognition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010357352.8A CN111524151A (en) | 2020-04-29 | 2020-04-29 | Object detection method and system based on background recognition |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111524151A true CN111524151A (en) | 2020-08-11 |
Family
ID=71905547
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010357352.8A Pending CN111524151A (en) | 2020-04-29 | 2020-04-29 | Object detection method and system based on background recognition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111524151A (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5065237A (en) * | 1990-08-01 | 1991-11-12 | General Electric Company | Edge detection using patterned background |
JPH07271989A (en) * | 1994-03-31 | 1995-10-20 | Toshiba Corp | Boundary detecting method of background and object area, contour extracting method of object and object contour extracting device |
DE19607258A1 (en) * | 1996-02-27 | 1997-08-28 | Olaf Haehnel | Object size and/or position determination device |
JP2015001835A (en) * | 2013-06-14 | 2015-01-05 | セコム株式会社 | Image sensor |
CN105976396A (en) * | 2016-04-27 | 2016-09-28 | 梧州市自动化技术研究开发院 | Method used for identifying moving object on conveyor belt in video image |
CN106408633A (en) * | 2016-05-09 | 2017-02-15 | 深圳市微蜂机器认知技术有限公司 | Object imaging method |
CN207397296U (en) * | 2017-11-14 | 2018-05-22 | 湖南大学 | A kind of object identification device based on figure viewed from behind imaging |
-
2020
- 2020-04-29 CN CN202010357352.8A patent/CN111524151A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5065237A (en) * | 1990-08-01 | 1991-11-12 | General Electric Company | Edge detection using patterned background |
JPH07271989A (en) * | 1994-03-31 | 1995-10-20 | Toshiba Corp | Boundary detecting method of background and object area, contour extracting method of object and object contour extracting device |
DE19607258A1 (en) * | 1996-02-27 | 1997-08-28 | Olaf Haehnel | Object size and/or position determination device |
JP2015001835A (en) * | 2013-06-14 | 2015-01-05 | セコム株式会社 | Image sensor |
CN105976396A (en) * | 2016-04-27 | 2016-09-28 | 梧州市自动化技术研究开发院 | Method used for identifying moving object on conveyor belt in video image |
CN106408633A (en) * | 2016-05-09 | 2017-02-15 | 深圳市微蜂机器认知技术有限公司 | Object imaging method |
CN207397296U (en) * | 2017-11-14 | 2018-05-22 | 湖南大学 | A kind of object identification device based on figure viewed from behind imaging |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102103206B (en) | Vehicle monitoring system | |
KR101817819B1 (en) | Method and device for monitoring the surroundings of a vehicle | |
US10545228B2 (en) | Object identification device | |
EP1821117B1 (en) | A method for processing an echo profile and a pulse-echo ranging system | |
WO2023024493A1 (en) | Signal processing method and detection method for lidar, and lidar | |
US10234548B2 (en) | Ultrasonic detection device to determine interference source by an additional reception mode | |
CN108303690B (en) | Ranging method and ranging system for eliminating laser radar blind area | |
CN110889974B (en) | Intelligent parking space identification method and device and automobile | |
WO2005081792A3 (en) | Method, apparatus and program for detecting an object | |
US20210063148A1 (en) | Method and device for operating an acoustic sensor | |
WO2023125322A2 (en) | Lidar echo signal processing method and apparatus, and computer device | |
MX2022001301A (en) | Information-enhanced off-vehicle event identification. | |
CN109263649B (en) | Vehicle, object recognition method and object recognition system thereof in automatic driving mode | |
CN111524151A (en) | Object detection method and system based on background recognition | |
US20210003676A1 (en) | System and method | |
US20200025916A1 (en) | Method and device for providing ultrasonic signal information | |
US11892569B2 (en) | Method and device for optical distance measurement | |
JP3647696B2 (en) | Turbulent bed detector | |
US10732255B2 (en) | Method and apparatus for detection of a signal | |
WO2023019573A1 (en) | Ranging method, waveform detection method, apparatus, and related device | |
US11789145B2 (en) | Method and apparatus for detecting at least partial flooding of a motor vehicle | |
CN109814106B (en) | Automobile safety early warning method | |
US11209541B1 (en) | Estimating a location of an object in close proximity to an ultrasonic transducer | |
US10480932B2 (en) | Automated computation of a dimension of a moving platform | |
CN109001712B (en) | Infrared distance sensor calibration method and intelligent household appliance |
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 |