CN111914670A - Method, device and system for detecting left-over article and storage medium - Google Patents
Method, device and system for detecting left-over article and storage medium Download PDFInfo
- Publication number
- CN111914670A CN111914670A CN202010650659.7A CN202010650659A CN111914670A CN 111914670 A CN111914670 A CN 111914670A CN 202010650659 A CN202010650659 A CN 202010650659A CN 111914670 A CN111914670 A CN 111914670A
- Authority
- CN
- China
- Prior art keywords
- feature points
- area
- region
- suspicious
- preset value
- 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
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000003860 storage Methods 0.000 title claims abstract description 16
- 238000001514 detection method Methods 0.000 claims abstract description 71
- 230000009466 transformation Effects 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- 238000005286 illumination Methods 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 12
- 238000004590 computer program Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 230000008859 change Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/20—Scenes; Scene-specific elements in augmented reality scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/757—Matching configurations of points or features
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The application discloses a method, a device, a system and a storage medium for detecting a left article, which are used for avoiding false detection of the left article due to changes of weather and illumination. The embodiment of the application provides a method for detecting a left-over article, which comprises the following steps: establishing a background model; acquiring a motion area in a detection image, and extracting a first characteristic point in the motion area; acquiring a previous frame image of the detection image, and extracting a second characteristic point from an area which is overlapped with the motion area in the previous frame image; matching the first characteristic points with the second characteristic points, and determining the number of matched characteristic points in the first characteristic points and the second characteristic points; judging whether the ratio of the number of the matched feature points to the number of the first feature points or the second feature points is smaller than a first preset value or not; and when the ratio of the number of the matched feature points to the number of the first feature points or the second feature points is smaller than a first preset value, taking the moving area as a first suspicious area, and judging whether the first suspicious area comprises the left-over article.
Description
Technical Field
The present application relates to the field of image processing technologies, and in particular, to a method, an apparatus, a system, and a storage medium for detecting a left-behind object.
Background
With the urbanization construction, the personnel distribution tends to be dense, and the public safety gradually becomes the focus of national government attention; terrorists leave unknown dangerous goods in places such as stations, subways and the like to make terrorist events in many countries and regions; although the explosion-proof and terrorist-proof device is arranged in places such as a station subway, how to effectively discover unknown left articles and give an alarm in time becomes a great problem of effectively solving possible hidden dangers.
In recent years, network cameras are deployed in a large scale, security and protection facilities are continuously perfected, and functions of an intelligent monitoring system are gradually expanded; the detection of the left-over articles becomes a basic requirement for camera monitoring in public places; the detection of the left-over articles is not only in the occasions with dense and large mobility, such as stations, subways and the like, but also has important significance in the aspects of corridor blockage, fire fighting access occupation detection and the like; at present, detection of the left-over articles is mainly based on two types of methods: the first type adopts a training-based method to train and detect various targets which are possibly left, and the method obviously has the defect of missing detection; the second type adopts background modeling and background updating methods, and is distinguished through target pixel values and different color space characteristics, and the methods are easily influenced by factors such as weather and light change.
In summary, the method for detecting the left-over article in the prior art has the problem of missing detection or false detection.
Disclosure of Invention
The embodiment of the application provides a method, a device and a system for detecting a left article and a storage medium, which are used for avoiding false detection of the left article due to changes of weather and illumination.
The embodiment of the application provides a method for detecting a legacy article, which comprises the following steps:
acquiring a background frame image, and establishing a background model according to the background frame image;
acquiring a motion area in a detection image according to the background model, and extracting a first feature point in the motion area;
acquiring a previous frame image of the detection image, and extracting a second feature point in an area which is overlapped with the motion area in the previous frame image;
matching the first feature points and the second feature points, and determining the number of matched feature points in the first feature points and the second feature points;
judging whether the ratio of the number of the matched feature points to the number of the first feature points or the second feature points is smaller than a first preset value or not;
and when the ratio of the number of the matched feature points to the number of the first feature points or the second feature points is smaller than a first preset value, taking the moving area as a first suspicious area, and judging whether the first suspicious area comprises a left article.
Optionally, extracting a first feature point in the motion region specifically includes:
extracting scale-invariant feature transformation feature points in the motion region as the first feature points;
extracting a second feature point from a region in the previous frame image coinciding with the motion region, specifically comprising:
and extracting scale-invariant feature transformation feature points in the overlapped region as the second feature points.
Optionally, matching the first feature point and the second feature point specifically includes:
calculating a Euclidean distance between the first feature point and the second feature point;
and taking the first characteristic point and the second characteristic point of which the Euclidean distance is smaller than a second preset value as matching characteristic points.
Optionally, determining whether the first suspicious region includes a legacy article specifically includes:
determining a pedestrian region in the detection image;
acquiring an intersection region of the pedestrian region and the first suspicious region;
judging whether the ratio of the intersection area to the first suspicious area is larger than or equal to a third preset value or not;
when the ratio of the intersection area to the first suspicious area is greater than or equal to the third preset value, performing background updating on the first suspicious area;
when the ratio of the intersection area to the first suspicious area is smaller than a third preset value, taking the intersection area as a second suspicious area;
judging whether the number of frames existing in the second suspicious region is greater than a fourth preset value;
when the number of frames of the second suspicious region is less than or equal to a fourth preset value, performing background updating on the first suspicious region;
and when the number of the frames existing in the second suspicious region is larger than a fourth preset value, determining that the second suspicious region comprises the left-over articles.
Optionally, determining the pedestrian region in the detection image specifically includes:
and detecting the detection image by using a pre-trained pedestrian model, and determining a pedestrian region in the detection image.
Optionally, establishing a background model according to the background frame image specifically includes:
and establishing a background model by adopting a single Gaussian background modeling method according to the background frame image.
Optionally, the method further comprises:
and when the ratio of the number of the matched characteristic points to the number of the first characteristic points or the second characteristic points is greater than or equal to a first preset value, updating the motion area.
The embodiment of the application provides a leave-on article detection device, leave-on article detection device includes:
the background modeling module is used for acquiring a background frame image and establishing a background model according to the background frame image;
the characteristic point extraction module is used for acquiring a motion area in a detection image according to the background model, extracting a first characteristic point in the motion area, acquiring a previous frame image of the detection image, and extracting a second characteristic point in an area which is overlapped with the motion area in the previous frame image;
the feature point matching module is used for matching the first feature points with the second feature points and determining the number of matched feature points in the first feature points and the second feature points;
and the judging module is used for judging whether the ratio of the number of the matched feature points to the number of the first feature points or the number of the second feature points is smaller than a first preset value, and when the ratio of the number of the matched feature points to the number of the first feature points or the number of the second feature points is smaller than the first preset value, the moving area is taken as a first suspicious area, and whether a left article is included in the first suspicious area is judged.
An embodiment of the present application provides a legacy article detection system, legacy article detection system includes:
a memory for storing program instructions;
and the processor is used for calling the program instructions stored in the memory and executing the steps included in the legacy article detection method provided by the embodiment of the application according to the obtained program instructions.
The storage medium stores computer-executable instructions, and the computer-executable instructions are used for causing a computer to execute the steps included in the legacy article detection method provided by the embodiment of the application.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic diagram of a method for detecting a left-behind object according to an embodiment of the present disclosure;
fig. 2 is a schematic diagram of another method for detecting a left-behind object according to an embodiment of the present disclosure;
fig. 3 is a schematic diagram of a legacy article detection device according to an embodiment of the present application;
fig. 4 is a schematic diagram of a legacy article detection system according to an embodiment of the present application.
Detailed Description
An embodiment of the present application provides a method for detecting a legacy article, as shown in fig. 1, the method includes:
s101, obtaining a background frame image, and establishing a background model for the background frame image;
s102, obtaining a motion area in a detection image according to the background model, and extracting a first feature point in the motion area;
s103, acquiring a previous frame image of the detection image, and extracting a second feature point in an area, which is overlapped with the motion area, in the previous frame image;
s104, matching the first feature points and the second feature points, and determining the number of matched feature points in the first feature points and the second feature points;
s105, judging whether the ratio of the number of the matched feature points to the number of the first feature points or the second feature points is smaller than a first preset value or not;
s106, when the ratio of the number of the matched feature points to the number of the first feature points or the second feature points is smaller than a first preset value, taking the motion area as a first suspicious area, and judging whether the first suspicious area comprises a legacy article.
According to the method for detecting the left-over article, the first characteristic point and the second characteristic point are respectively extracted from the motion area in the detected image and the area which is overlapped with the motion area in the previous frame of image, the first characteristic point and the second characteristic point are matched, the number of the matched characteristic points in the first characteristic point and the second characteristic point is determined, only when the ratio of the number of the matched characteristic points to the number of the first characteristic point or the second characteristic point is smaller than a first preset value, the motion area is taken as a first suspicious area, whether the motion area comprises the left-over article or not is judged, therefore, the false detection of the left-over article due to weather and illumination change can be effectively avoided, and the accuracy of the detection of the left-over article is improved.
Optionally, the method further comprises:
and when the ratio of the number of the matched characteristic points to the number of the first characteristic points or the second characteristic points is greater than or equal to a first preset value, updating the motion area.
When the ratio of the number of the matched characteristic points to the number of the first characteristic points or the second characteristic points is greater than or equal to a first preset value, the moving area is considered to be a normal area without the left articles, and background updating can be carried out on the moving area.
In specific implementation, the first preset value may be selected according to actual needs, and the first preset value may be, for example, 0.6.
In specific implementation, optionally, step S101 establishes a background model according to the background frame image, and specifically includes:
and establishing a background model by adopting a single Gaussian background modeling method according to the background frame image.
Optionally, in step S102, extracting a first feature point in the motion region specifically includes:
extracting Scale-invariant feature transform (SIFT) feature points in the motion region as the first feature points;
in step S103, extracting a second feature point from a region in the previous frame image that coincides with the motion region specifically includes:
and extracting SIFT feature points in the overlapped region as the second feature points.
Namely, the method for detecting the left-over article provided by the embodiment of the application adopts an SIFT algorithm, and the feature points are extracted from the motion area of the detected image and the area overlapped with the motion area in the previous frame image.
In specific implementation, an SIFT algorithm is adopted to search extreme points in a scale space and extract position, scale and direction information of the extreme points. The feature points extracted by the SIFT algorithm have good stability and invariance, can adapt to rotation, scale scaling and brightness change, and can be free from the interference of view angle change, affine transformation and noise to a certain extent. Therefore, the situation that the moving area is judged to be a suspicious area by mistake due to changes of weather and illumination can be effectively avoided, the false detection of the left articles is avoided, and the accuracy rate of the detection of the left articles is improved.
In specific implementation, the first feature point and the second feature point obtained by adopting the SIFT algorithm are feature points with 128-dimensional feature vectors.
Optionally, in step S104, matching the first feature point and the second feature point specifically includes:
calculating a Euclidean distance between the first feature point and the second feature point;
and taking the first characteristic point and the second characteristic point of which the Euclidean distance is smaller than a second preset value as matching characteristic points.
The smaller the Euclidean distance between the first feature point and the second feature point is, the higher the matching degree is. In a specific implementation, the second preset value may be, for example.
Optionally, in step S106, determining whether the first suspicious region includes a legacy article specifically includes:
determining a pedestrian region in the detection image;
acquiring an intersection region of the pedestrian region and the first suspicious region;
judging whether the ratio of the intersection area to the first suspicious area is larger than or equal to a third preset value or not;
when the ratio of the intersection area to the first suspicious area is greater than or equal to the third preset value, performing background updating on the first suspicious area;
when the ratio of the intersection area to the first suspicious area is smaller than a third preset value, taking the intersection area as a second suspicious area;
judging whether the number of frames existing in the second suspicious region is greater than a fourth preset value;
when the number of frames of the second suspicious region is less than or equal to a fourth preset value, performing background updating on the first suspicious region;
and when the number of the frames existing in the second suspicious region is larger than a fourth preset value, determining that the second suspicious region comprises the left-over articles.
The method for detecting the left-behind object provided by the embodiment of the application detects the pedestrian region, compares the pedestrian region with the first suspicious region, can avoid misjudging the pedestrian region into the left-behind object region, can further avoid the false detection of the left-behind object, and improves the detection accuracy.
In specific implementation, the third preset value and the fourth preset value can be selected according to actual needs. The third preset value may be, for example, 0.2. The fourth preset value may be, for example, 25.
Optionally, determining the pedestrian region in the detection image specifically includes:
and detecting the detection image by using a pre-trained pedestrian model, and determining a pedestrian region in the detection image.
In specific implementation, the pedestrian model trained in advance can be obtained by utilizing pedestrian images of monitoring scenes such as station subways and the like obtained in advance and adopting a deep learning network for training. The deep learning network is adopted for training, for example, the pedestrian in the pedestrian graph can be subjected to rectangle labeling, and the labeled data is trained by adopting 40 convolutional layers and 5 downsampling layers.
Next, taking the detection of the pedestrian area as an example, the method for detecting the left-behind article provided by the embodiment of the present application is exemplified, as shown in fig. 2, the method for detecting the left-behind article includes the following steps:
s201, obtaining a background frame image, and establishing a background model according to the background frame image;
s202, acquiring a motion region in the detection image, and extracting SIFT feature points from the motion region as first feature points;
s203, acquiring a previous frame image of the detection image, and extracting SIFT feature points from a region which is overlapped with the motion region in the previous frame image to serve as second feature points;
s204, matching the first characteristic point with the second characteristic point;
s205, judging whether the ratio of the number of the matched feature points to the number of the first feature points or the second feature points is smaller than a first preset value or not; if yes, go to step 207, otherwise go to step 206;
s206, updating the background model of the motion area;
s207, taking the motion area as a first suspicious area;
s208, determining a pedestrian area in the detected image;
s209, acquiring an intersection region of the pedestrian region and the first suspicious region;
s210, judging whether the ratio of the intersection area to the first suspicious area is larger than or equal to a third preset value or not; if yes, go to step S211, otherwise go to step 206;
s211, taking the intersection region as a second suspicious region;
s212, judging whether the number of frames in the second suspicious region is larger than a fourth preset value; if yes, go to step S213, otherwise go to step 206;
s213, determining that the second suspicious region comprises the legacy article.
Based on the same inventive concept, an embodiment of the present application further provides a legacy article detection apparatus, as shown in fig. 3, the legacy article detection apparatus includes:
the background modeling module 301 is configured to obtain a background frame image and establish a background model according to the background frame image;
a feature point extraction module 302, configured to obtain a motion region in a detection image according to the background model, and extract a first feature point in the motion region, and obtain a previous frame image of the detection image, and extract a second feature point in a region in the previous frame image that coincides with the motion region;
a feature point matching module 303, configured to match the first feature point with the second feature point, and determine the number of matched feature points in the first feature point and the second feature point;
the determining module 304 determines whether a ratio of the number of the matched feature points to the number of the first feature points or the number of the second feature points is smaller than a first preset value, and when the ratio of the number of the matched feature points to the number of the first feature points or the number of the second feature points is smaller than the first preset value, takes the moving area as a first suspicious area, and determines whether the first suspicious area includes a left-over article.
Optionally, the determining module is further configured to: and when the ratio of the number of the matched characteristic points to the number of the first characteristic points or the second characteristic points is greater than or equal to a first preset value, updating the motion area.
Optionally, the background modeling module is specifically configured to:
and establishing a background model by adopting a single Gaussian background modeling method according to the background frame image.
Optionally, the feature point extracting module is configured to extract a first feature point in the motion region, and specifically includes:
extracting SIFT feature points from the motion region as the first feature points;
the feature point extraction module is configured to extract a second feature point in a region in the previous frame image that coincides with the motion region, and specifically includes:
and extracting SIFT feature points in the overlapped region as the second feature points.
Optionally, the feature point matching module is configured to match the first feature point with the second feature point, and specifically includes: calculating a Euclidean distance between the first feature point and the second feature point;
and taking the first characteristic point and the second characteristic point of which the Euclidean distance is smaller than a second preset value as matching characteristic points.
Optionally, the legacy article detection device further comprises:
the pedestrian detection module is used for determining a pedestrian region in the detection image and acquiring an intersection region of the pedestrian region and the first suspicious region;
the judging module is also used for:
judging whether the ratio of the intersection area to the first suspicious area is larger than or equal to a third preset value or not;
when the ratio of the intersection area to the first suspicious area is greater than or equal to the third preset value, performing background updating on the first suspicious area;
when the ratio of the intersection area to the first suspicious area is smaller than a third preset value, taking the intersection area as a second suspicious area;
judging whether the number of frames existing in the second suspicious region is greater than a fourth preset value;
when the number of frames of the second suspicious region is less than or equal to a fourth preset value, performing background updating on the first suspicious region;
and when the number of the frames existing in the second suspicious region is larger than a fourth preset value, determining that the second suspicious region comprises the left-over articles.
Optionally, the pedestrian detection module is specifically configured to:
and detecting the detection image by using a pre-trained pedestrian model, and determining a pedestrian region in the detection image.
Based on the same inventive concept, an embodiment of the present application further provides a legacy article detection system, as shown in fig. 4, the legacy article detection system includes:
a memory for storing program instructions;
and the processor is used for calling the program instructions stored in the memory and executing the steps included in the legacy article detection method provided by the embodiment of the application according to the obtained program instructions.
In particular implementation, as shown in fig. 4, the legacy article detection system provided by the embodiment of the present application may include, for example, at least one processor 402, and a memory 401 connected to the at least one processor 402. In the embodiment of the present application, a specific connection medium between the processor 402 and the memory 401 is not limited, fig. 4 is an example in which the processor 402 and the memory 401 are connected by a bus 400, the bus 400 is represented by a thick line in fig. 4, and a connection manner between other components is merely schematically illustrated and is not limited thereto. The bus 400 may be divided into an address bus, a data bus, a control bus, etc., and is shown with only one thick line in fig. 4 for ease of illustration, but does not represent only one bus or type of bus.
In the system for detecting a legacy article provided by the embodiment of the present application, the memory 401 stores instructions executable by the at least one processor 402, and the at least one processor 402 may execute the steps included in the method for detecting a legacy article provided by the embodiment of the present application by calling the instructions stored in the memory 401.
The processor 402 is a control center of the legacy article detection system, and can utilize various interfaces and lines to connect various parts of the whole legacy article detection system, and implement various functions of the legacy article detection system by executing instructions stored in the memory 401. Optionally, the processor 402 may include one or more processing units, and the processor 402 may integrate an application processor and a modem processor, wherein the application processor mainly handles operating systems, user interfaces, application programs, and the like, and the modem processor mainly handles wireless communications. It will be appreciated that the modem processor described above may not be integrated into the processor 402. In some embodiments, processor 402 and memory 401 may be implemented on the same chip, or in some embodiments, they may be implemented separately on separate chips.
The processor 402 may be a general-purpose processor, such as a Central Processing Unit (CPU), digital signal processor, application specific integrated circuit, field programmable gate array or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or the like, that may implement or perform the methods, steps, and logic blocks disclosed in embodiments of the present application. A general purpose processor may be a microprocessor or any conventional processor or the like. The steps of the method for detecting a legacy object disclosed in the embodiments of the present application may be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor.
By programming the processor 402, the code corresponding to the method for detecting a left-over object described in the foregoing embodiment may be solidified into a chip, so that the chip can execute the steps of the method for detecting a left-over object when running, and how to program the processor 402 is a technique known by those skilled in the art and will not be described herein again.
Based on the same inventive concept, the embodiment of the present application further provides a storage medium, where the storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the steps included in the legacy article detection method provided by the embodiment of the present application.
To sum up, according to the method, the apparatus, the system, and the storage medium for detecting a left article provided in the embodiment of the present application, the first feature point and the second feature point are respectively extracted from the motion region in the detection image and the region coinciding with the motion region in the previous frame image, and the first feature point and the second feature point are matched to determine the number of matched feature points in the first feature point and the second feature point, and only when the ratio of the number of matched feature points to the number of first feature points or the number of second feature points is smaller than a first preset value, the motion region is used as a first suspicious region, and whether a left article is included in the motion region is determined, so that false detection of the left article due to weather and illumination changes can be effectively avoided, and the accuracy of detecting the left article is improved.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.
Claims (10)
1. A method for detecting a carry-over, the method comprising:
acquiring a background frame image, and establishing a background model according to the background frame image;
acquiring a motion area in a detection image according to the background model, and extracting a first feature point in the motion area;
acquiring a previous frame image of the detection image, and extracting a second feature point in an area which is overlapped with the motion area in the previous frame image;
matching the first feature points and the second feature points, and determining the number of matched feature points in the first feature points and the second feature points;
judging whether the ratio of the number of the matched feature points to the number of the first feature points or the second feature points is smaller than a first preset value or not;
and when the ratio of the number of the matched feature points to the number of the first feature points or the second feature points is smaller than a first preset value, taking the moving area as a first suspicious area, and judging whether the first suspicious area comprises a left article.
2. The method according to claim 1, wherein extracting the first feature point in the motion region specifically comprises:
extracting scale-invariant feature transformation feature points in the motion region as the first feature points;
extracting a second feature point from a region in the previous frame image coinciding with the motion region, specifically comprising:
and extracting scale-invariant feature transformation feature points in the overlapped region as the second feature points.
3. The method according to claim 2, wherein matching the first feature point and the second feature point specifically comprises:
calculating a Euclidean distance between the first feature point and the second feature point;
and taking the first characteristic point and the second characteristic point of which the Euclidean distance is smaller than a second preset value as matching characteristic points.
4. The method according to claim 1, wherein determining whether the first suspect region includes a legacy item comprises:
determining a pedestrian region in the detection image;
acquiring an intersection region of the pedestrian region and the first suspicious region;
judging whether the ratio of the intersection area to the first suspicious area is larger than or equal to a third preset value or not;
when the ratio of the intersection area to the first suspicious area is greater than or equal to the third preset value, performing background updating on the first suspicious area;
when the ratio of the intersection area to the first suspicious area is smaller than a third preset value, taking the intersection area as a second suspicious area;
judging whether the number of frames existing in the second suspicious region is greater than a fourth preset value;
when the number of frames of the second suspicious region is less than or equal to a fourth preset value, performing background updating on the first suspicious region;
and when the number of the frames existing in the second suspicious region is larger than a fourth preset value, determining that the second suspicious region comprises the left-over articles.
5. The method according to claim 4, wherein determining the pedestrian region in the detection image specifically comprises:
and detecting the detection image by using a pre-trained pedestrian model, and determining a pedestrian region in the detection image.
6. The method according to claim 1, wherein building a background model from the background frame image specifically comprises:
and establishing a background model by adopting a single Gaussian background modeling method according to the background frame image.
7. The method of claim 1, further comprising:
and when the ratio of the number of the matched characteristic points to the number of the first characteristic points or the second characteristic points is greater than or equal to a first preset value, updating the motion area.
8. A left-behind item detecting apparatus, characterized in that the left-behind item detecting apparatus comprises:
the background modeling module is used for acquiring a background frame image and establishing a background model according to the background frame image;
the characteristic point extraction module is used for acquiring a motion area in a detection image according to the background model, extracting a first characteristic point in the motion area, acquiring a previous frame image of the detection image, and extracting a second characteristic point in an area which is overlapped with the motion area in the previous frame image;
the feature point matching module is used for matching the first feature points with the second feature points and determining the number of matched feature points in the first feature points and the second feature points;
and the judging module is used for judging whether the ratio of the number of the matched feature points to the number of the first feature points or the number of the second feature points is smaller than a first preset value, and when the ratio of the number of the matched feature points to the number of the first feature points or the number of the second feature points is smaller than the first preset value, the moving area is taken as a first suspicious area, and whether a left article is included in the first suspicious area is judged.
9. A legacy article detection system, comprising:
a memory for storing program instructions;
a processor for calling said program instructions stored in said memory and executing the steps comprised by the method according to any one of claims 1 to 7 according to said obtained program instructions.
10. A storage medium storing computer-executable instructions for causing a computer to perform the steps comprising the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010650659.7A CN111914670A (en) | 2020-07-08 | 2020-07-08 | Method, device and system for detecting left-over article and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010650659.7A CN111914670A (en) | 2020-07-08 | 2020-07-08 | Method, device and system for detecting left-over article and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111914670A true CN111914670A (en) | 2020-11-10 |
Family
ID=73226425
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010650659.7A Pending CN111914670A (en) | 2020-07-08 | 2020-07-08 | Method, device and system for detecting left-over article and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111914670A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112668423A (en) * | 2020-12-18 | 2021-04-16 | 平安科技（深圳）有限公司 | Corridor sundry detection method and device, terminal equipment and storage medium |
| CN113065425A (en) * | 2021-03-19 | 2021-07-02 | 东风汽车集团股份有限公司 | Method and system for reminding objects left in vehicle based on environmental information and storage medium |
| CN113470013A (en) * | 2021-07-28 | 2021-10-01 | 浙江大华技术股份有限公司 | Method and device for detecting moved article |
| CN117152751A (en) * | 2023-10-30 | 2023-12-01 | 西南石油大学 | Image segmentation method and system |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20090044957A (en) * | 2007-10-30 | 2009-05-07 | 삼성전자주식회사 | Theft and left baggage survellance system and meothod thereof |
| CN102063614A (en) * | 2010-12-28 | 2011-05-18 | 天津市亚安科技电子有限公司 | Method and device for detecting lost articles in security monitoring |
| CN102509075A (en) * | 2011-10-19 | 2012-06-20 | 北京国铁华晨通信信息技术有限公司 | Remnant object detection method and device |
| CN104156942A (en) * | 2014-07-02 | 2014-11-19 | 华南理工大学 | Detection method for remnants in complex environment |
| CN104809742A (en) * | 2015-04-15 | 2015-07-29 | 广西大学 | Article safety detection method in complex scene |
| CN105590086A (en) * | 2014-11-17 | 2016-05-18 | 西安三茗科技有限责任公司 | Article antitheft detection method based on visual tag identification |
| CN107527009A (en) * | 2017-07-11 | 2017-12-29 | 浙江汉凡软件科技有限公司 | A kind of remnant object detection method based on YOLO target detections |
| US20190095745A1 (en) * | 2017-09-28 | 2019-03-28 | Baidu Usa Llc | Systems and methods to improve visual feature detection using motion-related data |
| CN110348327A (en) * | 2019-06-24 | 2019-10-18 | 腾讯科技（深圳）有限公司 | Realize the method and device that Articles detecting is left in monitoring scene |
-
2020
- 2020-07-08 CN CN202010650659.7A patent/CN111914670A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20090044957A (en) * | 2007-10-30 | 2009-05-07 | 삼성전자주식회사 | Theft and left baggage survellance system and meothod thereof |
| CN102063614A (en) * | 2010-12-28 | 2011-05-18 | 天津市亚安科技电子有限公司 | Method and device for detecting lost articles in security monitoring |
| CN102509075A (en) * | 2011-10-19 | 2012-06-20 | 北京国铁华晨通信信息技术有限公司 | Remnant object detection method and device |
| CN104156942A (en) * | 2014-07-02 | 2014-11-19 | 华南理工大学 | Detection method for remnants in complex environment |
| CN105590086A (en) * | 2014-11-17 | 2016-05-18 | 西安三茗科技有限责任公司 | Article antitheft detection method based on visual tag identification |
| CN104809742A (en) * | 2015-04-15 | 2015-07-29 | 广西大学 | Article safety detection method in complex scene |
| CN107527009A (en) * | 2017-07-11 | 2017-12-29 | 浙江汉凡软件科技有限公司 | A kind of remnant object detection method based on YOLO target detections |
| US20190095745A1 (en) * | 2017-09-28 | 2019-03-28 | Baidu Usa Llc | Systems and methods to improve visual feature detection using motion-related data |
| CN110348327A (en) * | 2019-06-24 | 2019-10-18 | 腾讯科技（深圳）有限公司 | Realize the method and device that Articles detecting is left in monitoring scene |
Non-Patent Citations (2)
| Title |
|---|
| 周宏 等: "一种基于目标检测的快速视频取证方法", 计算机应用研究, no. 03, 31 March 2009 (2009-03-31), pages 1089 - 1091 * |
| 屈凡非 等: "基于背景配准的矿井危险区域视频目标检测算法", 工矿自动化, no. 03, 28 February 2011 (2011-02-28), pages 48 - 50 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112668423A (en) * | 2020-12-18 | 2021-04-16 | 平安科技（深圳）有限公司 | Corridor sundry detection method and device, terminal equipment and storage medium |
| CN112668423B (en) * | 2020-12-18 | 2024-05-28 | 平安科技（深圳）有限公司 | Corridor sundry detection method and device, terminal equipment and storage medium |
| CN113065425A (en) * | 2021-03-19 | 2021-07-02 | 东风汽车集团股份有限公司 | Method and system for reminding objects left in vehicle based on environmental information and storage medium |
| CN113470013A (en) * | 2021-07-28 | 2021-10-01 | 浙江大华技术股份有限公司 | Method and device for detecting moved article |
| CN117152751A (en) * | 2023-10-30 | 2023-12-01 | 西南石油大学 | Image segmentation method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111914670A (en) | Method, device and system for detecting left-over article and storage medium | |
| Lin et al. | Abandoned object detection via temporal consistency modeling and back-tracing verification for visual surveillance | |
| Lim et al. | Background subtraction using encoder-decoder structured convolutional neural network | |
| CN107844794B (en) | Image recognition method and device | |
| US8744125B2 (en) | Clustering-based object classification | |
| US20170213081A1 (en) | Methods and systems for automatically and accurately detecting human bodies in videos and/or images | |
| US20100322475A1 (en) | Object area detecting device, object area detecting system, object area detecting method and program | |
| JP2001175876A (en) | Method and device for detecting moving body | |
| CN114119676B (en) | Target detection tracking identification method and system based on multi-feature information fusion | |
| Lee et al. | Context and profile based cascade classifier for efficient people detection and safety care system | |
| Jo | Cumulative dual foreground differences for illegally parked vehicles detection | |
| WO2012022744A2 (en) | Multi-mode video event indexing | |
| Han et al. | Moving object detection revisited: Speed and robustness | |
| Choi et al. | Multiple vehicles detection and tracking based on scale-invariant feature transform | |
| CN110502977B (en) | Building change classification detection method, system, device and storage medium | |
| Lin et al. | Particle filter with occlusion handling for visual tracking | |
| Shyam et al. | Abandoned object detection using pixel-based finite state machine and single shot multibox detector | |
| CN113012383A (en) | Fire detection alarm method, related system, related equipment and storage medium | |
| CN111914656A (en) | Personnel behavior detection method and device, electronic equipment and storage medium | |
| CN111797726A (en) | Flame detection method and device, electronic equipment and storage medium | |
| Parvin et al. | Vehicle number plate detection and recognition techniques: a review | |
| Filonenko et al. | Illegally parked vehicle detection using adaptive dual background model | |
| Zhang et al. | Image-based approach for parking-spot detection with occlusion handling | |
| CN112070033A (en) | Video carry-over detection method based on finite-state machine analysis | |
| Rahman et al. | A real-time end-to-end Bangladeshi license plate detection and recognition system for all situations including challenging environmental scenarios |
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 |