CN116416577A - Abnormality identification method for construction monitoring system - Google Patents
Abnormality identification method for construction monitoring system Download PDFInfo
- Publication number
- CN116416577A CN116416577A CN202310500967.5A CN202310500967A CN116416577A CN 116416577 A CN116416577 A CN 116416577A CN 202310500967 A CN202310500967 A CN 202310500967A CN 116416577 A CN116416577 A CN 116416577A
- Authority
- CN
- China
- Prior art keywords
- value
- abnormal
- frame difference
- speed
- 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.)
- Granted
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 96
- 238000010276 construction Methods 0.000 title claims abstract description 67
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000005856 abnormality Effects 0.000 title claims description 36
- 230000002159 abnormal effect Effects 0.000 claims abstract description 139
- 230000008859 change Effects 0.000 claims description 15
- 208000012661 Dyskinesia Diseases 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 5
- 238000001514 detection method Methods 0.000 abstract description 4
- 230000000694 effects Effects 0.000 description 6
- 230000000007 visual effect Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 206010000117 Abnormal behaviour Diseases 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000001965 increasing effect Effects 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000035484 reaction time Effects 0.000 description 1
- 230000035945 sensitivity Effects 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
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
-
- 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
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- 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
-
- 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
- G06T2207/10016—Video; Image sequence
-
- 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
- G06T2207/20212—Image combination
- G06T2207/20224—Image subtraction
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30232—Surveillance
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Geometry (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to an anomaly identification method for a construction monitoring system, which is used for acquiring a monitoring picture of a construction site; acquiring an information value of each pixel point based on a gray value and a gradient value of the pixel point in each frame of monitoring image; obtaining a corresponding frame difference image by calculating the information value difference of the pixel points at the corresponding positions of every two adjacent frames of monitoring images; identifying objects in the frame difference images, and performing corner matching on two adjacent frame difference images to obtain the moving distance of each object; acquiring a speed abnormal value and an area abnormal value of each object; obtaining an abnormal adjustment value; acquiring comprehensive abnormal degrees by combining the speed abnormal values, the area abnormal values, the abnormal adjustment values and the preset attenuation values of all objects in the frame difference image, and identifying that the construction site is abnormal when the comprehensive abnormal degrees are larger than a preset abnormal threshold value. The invention can reduce false detection of abnormal conditions of the construction site through image data processing and ensure construction safety.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an anomaly identification method for a construction monitoring system.
Background
The construction site condition is complicated and has more potential safety hazards, and some constructors have weak safety consciousness and cannot realize danger in time, so that irrecoverable loss is caused. The construction monitoring system composed of a plurality of monitoring devices and devices for displaying monitoring pictures can only acquire pictures of a construction site in real time and display the pictures, and can discover abnormal conditions of the construction site by watching monitoring videos artificially, so that the efficiency is low and the timeliness is poor.
With the development of science and technology, an intelligent construction monitoring system for timely finding dangerous situations of construction sites and identifying abnormal situations appears. The intelligent construction monitoring system generally builds an identification model by learning a large amount of data, and identifies abnormal behaviors in the acquired construction site images, but the method for identifying the abnormal behaviors requires a large amount of data to support, and is high in cost. Or whether the abnormality exists or not is judged by comparing the difference between the monitoring video image and the pre-acquired standard image, but the construction worker is not invariable in construction operation, and the abnormality is easy to be detected by mistake only by comparing the monitoring video image with the standard image, so that the abnormality identification is inaccurate.
Disclosure of Invention
In order to solve the problem of inaccurate abnormality identification of a construction monitoring system, the invention provides an abnormality identification method for the construction monitoring system, which adopts the following technical scheme:
acquiring a monitoring picture of a construction site, wherein the monitoring picture is a continuous monitoring image; acquiring an information value of each pixel point based on a gray value and a gradient value of the pixel point in each frame of monitoring image;
obtaining a corresponding frame difference image by calculating the information value difference of the pixel points at the corresponding positions of every two adjacent frames of monitoring images; identifying objects in the frame difference images, and performing corner matching on two adjacent frame difference images to obtain the moving distance of each object; acquiring a speed abnormal value of a corresponding object according to a time interval and a moving distance between each frame of frame difference image and an adjacent frame of frame difference image, and acquiring an area abnormal value based on the area of each object in the frame difference image;
acquiring a normal moving range by utilizing the area change of a frame difference image under a preset period, and acquiring an abnormal adjustment value according to the normal moving range; acquiring the abnormal degree of the corresponding frame difference image by combining the abnormal speed value and the abnormal area value of all objects in each frame of frame difference image with the abnormal adjustment value and the preset attenuation value;
and accumulating the abnormal degrees of all the frame difference images to obtain comprehensive abnormal degrees, and identifying that the construction site is abnormal when the comprehensive abnormal degrees are larger than a preset abnormal threshold value.
Further, the information value obtaining method includes:
and respectively normalizing the gray value and the gradient value of each pixel point, and carrying out weighted summation on the obtained two normalization results to obtain the information value of the corresponding pixel point.
Further, the method for acquiring the frame difference image comprises the following steps:
for two adjacent frames of monitoring images, subtracting the information value of the pixel point at the corresponding position in the previous frame of monitoring image from the information value of any pixel point in the next frame of monitoring image, and generating the frame difference image by taking the obtained absolute value of the difference value as the frame difference value at the pixel point and taking the frame difference value as the pixel value of the corresponding pixel point.
Further, the method for obtaining the moving distance comprises the following steps:
and acquiring coordinates of the centroid pixel points of the object in two adjacent frames of frame difference images through the corner matching result, and acquiring the moving distance of the centroid pixel points as the moving distance of the corresponding object according to the coordinates of the centroid pixel points in the two frames of frame difference images.
Further, the speed anomaly value obtaining method includes:
and obtaining the maximum standard speed of the construction site, taking the ratio of the moving distance of each object to the corresponding time interval as the moving speed of the object, when the moving speed is greater than the maximum standard speed, the object moves abnormally, calculating the speed difference between the moving speed and the maximum standard speed, and taking the ratio of the speed difference in the moving speed as the speed abnormal value.
Further, the method for obtaining the area anomaly value comprises the following steps:
when there is abnormal movement of the object, the area anomaly value is obtained as a result of the area evolution of the object in the frame difference image.
Further, the method for obtaining the abnormal adjustment value comprises the following steps:
and calculating the proportion of the normal moving range to the area of the monitoring image as the abnormal adjustment value.
Further, the method for obtaining the abnormality degree comprises the following steps:
and for each object in the frame difference image, when the object moves abnormally, calculating the sum of the speed abnormal value and the area abnormal value, multiplying the abnormal adjustment value to obtain an abnormal value, adjusting the abnormal value by using the preset attenuation value, and summing the abnormal values after the adjustment of all the objects moving abnormally in the frame difference image to obtain the abnormal degree of the corresponding frame difference image.
The invention has at least the following beneficial effects:
the information value of each pixel point is obtained by combining the gray value and the gradient value of the pixel point in each frame of monitoring image, and the information value which more obviously reflects the information of the pixel point is obtained by combining the gray value and the gradient value; then, obtaining a frame difference image by carrying out frame difference on the information value, enhancing the difference between two adjacent frame monitoring images, carrying out corner matching on the adjacent frame difference images to obtain the moving distance of the object, reflecting the moving condition of the object through the change of the same object in the adjacent frame difference images, further obtaining a speed abnormal value to judge the possibility of occurrence of moving abnormality, and reflecting the possibility of occurrence of area abnormality by the area of each frame difference image; further, a normal moving range is obtained according to the area change of the frame difference image under a preset period, a constructor usually moves only in a fixed range when working, the normal moving range of the constructor is reflected through the area change of the frame difference image under a period of time, an abnormal adjusting value is further obtained, the abnormal adjusting value is obtained under the condition that the normal moving range is known, then the abnormal degree of each frame of frame difference image is obtained by combining the abnormal speed value, the abnormal area value and the attenuation value, and the abnormal degree is obtained by combining various abnormal indexes; and then accumulating the abnormal degrees of all the frame difference images as the comprehensive abnormal degrees to perform abnormal recognition. According to the invention, the indexes reflecting the abnormality can be obtained in multiple aspects through data processing, so that accurate abnormality identification results are obtained by combining the indexes, false detection of the abnormal condition of the construction site is reduced, the construction safety is ensured, and the construction efficiency is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of steps of an anomaly identification method for a construction monitoring system according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purposes, the following detailed description refers to specific embodiments, structures, features and effects of an anomaly identification method for a construction monitoring system according to the present invention with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of an abnormality identification method for a construction monitoring system provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of steps of an anomaly identification method for a construction monitoring system according to an embodiment of the present invention is shown, the method includes the following steps:
step S001, obtaining a monitoring picture of a construction site, wherein the monitoring picture is a continuous monitoring image; and acquiring an information value of each pixel point based on the gray value and the gradient value of the pixel point in each frame of monitoring image.
The construction monitoring system comprises a monitoring camera and a monitoring center, can collect monitoring pictures of all construction sites in real time, and each monitoring picture corresponds to one monitoring visual angle and comprises continuous multi-frame monitoring images. The monitoring camera needs to have the highest picture quality, for example, a 400-ten thousand ultra-clean network ball machine is adopted.
The content in the pictures collected at the construction site comprises constructors, equipment and devices, building facilities, background such as ground sky and the like. In consideration of that a small amount of noise occurs in the monitoring picture due to signal fluctuation and the like, noise in the image is removed through median filtering in the embodiment, and noise interference is avoided. The median filtering is a well-known technique in the image processing field, and the specific process is not described in detail in this embodiment.
And for each frame of monitoring image, calculating the gradient value of each pixel point, obtaining the pixel value of each pixel point, respectively normalizing the gray value and the gradient value of each pixel point, and carrying out weighted summation on the obtained two normalization results to obtain the information value of the corresponding pixel point.
The normalized result of the pixel value of the nth pixel point is recorded as
Normalization of gradient valuesIs->
Then
And->
The value ranges of the (E) are all 0,1]The pixel value and the gradient value reflect the information of the pixel point, and the information value of the nth pixel point is obtained by weighting and summing the two normalization results>
. The brightness of the moving object changes due to the background information in the picture, such as the change of light, and the like, and the frame difference image can capture the change. Therefore, compared with the traditional frame difference calculation based on pixel values, the method takes the gradient values as main judgment basis, and meanwhile, the method mainly judges the abnormality through the movement of the object, so that the weight of the gradient values is slightly larger than the pixel values, and weights of 0.6 and 0.4 are respectively given to the gradient values and the pixel values in the embodiment. The information value obtained by combining the gray value and the gradient value can more obviously reflect the information of the pixel point.
Step S002, obtaining a corresponding frame difference image by calculating the difference of the information values of the pixel points at the corresponding positions of every two adjacent frames of monitoring images; identifying objects in the frame difference images, and performing corner matching on two adjacent frame difference images to obtain the moving distance of each object; and acquiring a speed abnormal value of a corresponding object according to the time interval and the moving distance between each frame of frame difference image and the adjacent frame of frame difference image, and acquiring an area abnormal value based on the area of each object in the frame difference image.
For two adjacent frames of monitoring images, subtracting the information value of the pixel point at the corresponding position in the previous frame of monitoring image from the information value of any pixel point in the next frame of monitoring image, and generating a frame difference image by taking the obtained absolute value of the difference as the frame difference at the pixel point and taking the frame difference as the pixel value of the corresponding pixel point.
Monitoring for adjacent ith frameTaking the nth pixel point as an example, the frame difference value obtained by calculation is as follows:
wherein->
Representing the frame difference value of the nth pixel point in the ith frame difference image, +.>
Information value representing the nth pixel point in the i+1th frame monitoring image, +.>
Information value representing the nth pixel point in the ith frame of monitoring image,/for example>
Representation->
And->
Is the absolute value of the difference of (c).
And generating a frame difference image by taking the frame difference value of each pixel point as a corresponding pixel value, and reflecting the movement condition of an object in the monitoring image, including the movement of equipment and the movement or action of constructors, through the information value difference of the adjacent two frames of monitoring images at the corresponding positions.
Because the constructor activity content brought by the fixed construction content is repeated, the corresponding change in the adjacent frame difference image is smaller, and the changed area size is also more continuous and distributed in the activity range. In the continuous frame difference picture, extra content is suddenly generated, and the emergency such as high-altitude falling objects or vehicles entering outside the monitoring picture is more likely. When these extra contents appear, there is a risk that the personal safety of the constructors in the monitoring image may be formed to be different in degree, and at this time, the subsequent abnormality is identified by the change of the frame difference image.
Specifically, firstly, identifying an object in the frame difference images and performing corner matching on two adjacent frame difference images. The method for identifying the object is connected domain analysis, and whether the connected pixel points occur in the corresponding eight neighborhood range is judged for each pixel point in the frame difference image through the connected domain analysis until all the connected pixel points form a connected domain, and each connected domain is an object.
And obtaining all the corner points of each object through corner point detection, then carrying out corner point matching on two adjacent frames of frame difference images through a cann algorithm, and judging the corresponding relation of the corresponding objects in the adjacent frame difference images through the matching result of each corner point. It should be noted that, the cann algorithm is a matching algorithm commonly used in the field of image data processing, and specific processes are not repeated in the embodiment of the present invention.
And acquiring coordinates of the centroid pixel points of the object in two adjacent frames of frame difference images through the corner matching result, and acquiring the moving distance of the centroid pixel points as the moving distance of the corresponding object according to the coordinates of the centroid pixel points in the two frames of frame difference images.
The method comprises the steps of obtaining the coordinates of mass center pixel points of each object in each frame of frame difference image, calculating the distance between the coordinates of mass center pixel points of two adjacent frames of frame difference images by utilizing a two-point distance formula to serve as the moving distance of a corresponding object, and recording the moving distance of a c-th object as the moving distance of the c-th object
。
And obtaining the maximum standard speed of the construction site, taking the ratio of the moving distance of each object to the corresponding time interval as the moving speed of the object, when the moving speed is greater than the maximum standard speed, the object moves abnormally, calculating the speed difference between the moving speed and the maximum standard speed, and taking the duty ratio of the speed difference in the moving speed as a speed abnormal value.
Obtaining the maximum speed generated by the construction site in a safety range as the corresponding maximum standard speed to be recorded as
Due to different construction scenes, the construction is correspondingly performedThe maximum standard speed acceptable in the scene is different, for example, the maximum standard speed of the high-altitude construction operation and the maximum standard speed of the ground operation can have large difference, so the maximum standard speed is +>
The values under different construction scenes are different, and the values need to be set according to actual conditions.
Calculating the movement speed of the c-th object
Wherein t represents a movement distance +.>
The interval time between two corresponding frame difference images. When the speed of movement is +.>
Greater than maximum specification speed->
When the object moves abnormally, calculating a speed difference between the moving speed and the maximum standard speed, and taking the ratio of the speed difference in the moving speed as a speed abnormal value:
representing that the movement speed of the c-th object exceeds the maximum specification speed +.>
The extent of (a) that is the moving speed and the maximum specification speed +.>
The greater the phase difference, the greater the speed anomaly and the greater the speed anomaly.
Objects that suddenly appear in the monitored screen, whether they be high-altitude objects or information such as vehicles that are coming in from outside the screen, are potentially dangerous to the person being constructed. And the danger degree of the suddenly appearing object can be visually and externally physically represented along with the speed, the volume and the like of the object. Whether the constructor is reminded to carry out early warning on the possible high-altitude falling objects or the large-size vehicle is reminded to enter and carefully avoid, the abnormal values are synchronously increased.
Thus, when there is abnormal movement of the object, the area anomaly value is calculated at the same time. Taking the result of the area evolution of the object in the frame difference image as an area anomaly value, i.e
Wherein->
An area anomaly value representing the c-th object, < >>
Representing the area of the c-th object in the frame difference image. Area of the c-th object in the frame difference image +.>
And counting the number of pixels of the c object in the frame difference image. The area anomaly value is obtained according to the area of the object in the frame difference image, the frame difference image reflects the abnormal situation of the object, the larger the area is, the more the object moves, and the purpose of the root mark is to adjust the value of the anomaly value corresponding to the area, so that the balance is prevented from being influenced by the overlarge value of the area anomaly value.
The degree of risk of an object appearing in the monitor screen can be represented by the speed abnormality value and the area abnormality value to increase with an increase in the speed and the volume of the object itself.
Step S003, acquiring a normal moving range by utilizing the area change of the frame difference image in a preset period, and acquiring an abnormal adjustment value according to the normal moving range; and acquiring the abnormal degree of the corresponding frame difference image by combining the abnormal speed value, the abnormal area value, the abnormal adjustment value and the preset attenuation value of all objects in each frame of frame difference image.
Considering the personal safety problem of constructors and the construction specification of a construction site, the corresponding moving path and speed are specified no matter the constructors move or the instruments move. This characteristic reflects that in successive frame difference pictures, the amplitude of the change between adjacent frames appears to be bounded. And in the construction monitoring screen, the change of the rest of the screen, i.e., the background content, is small, or even hardly occurs, except for the constructors and the equipment.
That is, under the same monitoring view angle, the construction contents of the constructors are fixed within a period of time, the repeatability is very strong, the fixed construction contents are represented as the fixed moving range of the constructors, and the area change range of the part which is reflected in the frame difference image and is moving is fixed.
Therefore, the normal moving range is obtained by utilizing the area change of the frame difference images in the preset time period, specifically, the continuous multi-frame difference images in the preset time period are obtained, the areas of the parts, which are not 0, of the pixel points in each frame of frame difference images are counted, and the average value of the areas in the frame difference images is calculated as the normal moving range.
The part with the pixel point not being 0 is the part with movement, and the average value of the area of the moving part in the preset time period is counted as the normal moving range, namely the normal moving range, and is recorded as
. In the embodiment of the invention, the preset time period is 10 minutes, that is, the average value of the areas of the parts, which are not 0, of the pixel points in all the frame difference images within 10 minutes is counted to be used as the normal activity range. In other embodiments, the duration of the preset period may be adjusted according to the actual situation.
And calculating the proportion of the normal moving range to the area of the monitoring image as an abnormal adjustment value. Namely, the calculation method of the anomaly adjustment value is as follows:
wherein->
Representing an abnormal adjustment value->
Indicates the normal activity range, +.>
Representing the monitored image area. Since the continuous multi-frame monitoring images are collected under the same monitoring visual angle, all the monitoring images used for calculating the normal moving range have the same size and are all +.>
。
The larger the proportion of the area of the normal moving range of the constructor in the monitoring picture is, the smaller the range of the peripheral area of the corresponding non-moving area is, and if an abnormal object appears in the peripheral area at the moment, the smaller the required reaction time is, the more dangerous is for the constructor, and the larger the abnormal adjustment value at the moment is, the larger the influence on the subsequent abnormal degree calculation is.
And for each object in the frame difference image, when the object moves abnormally, calculating the sum of the abnormal speed value and the abnormal area value, multiplying the abnormal speed value by the abnormal adjustment value to obtain an abnormal value, adjusting the abnormal value by using a preset attenuation value, and summing the abnormal values after the adjustment of all the objects moving abnormally in the frame difference image to obtain the abnormal degree of the corresponding frame difference image.
Taking the i-th frame difference image as an example, the degree of abnormality is calculated based on an object that may have abnormal movement:
wherein, the liquid crystal display device comprises a liquid crystal display device,
indicating the degree of abnormality of the i-th frame difference image,/->
Representing an abnormal adjustment value->
Representing the abnormal value of the area of the c-th object in the i-th frame difference image, +.>
Speed anomaly value representing the c-th object in the i-th frame difference image, +.>
Representing the attenuation value, C represents the total number of objects in the i-th frame difference image.
Attenuation value
For the preset, the function is to inhibit the unidirectional increase of the abnormal degree to a certain extent, and prevent the unrestricted superposition of abnormal values obtained by some weaker abnormal behaviors which occur under normal conditions. Attenuation values +.in the examples of the invention>
Preset to 20, in other embodiments the department of science adjusts attenuation values according to the actual situation
Is a value of (a).
When the moving speed is not greater than the maximum standard speed, i.e. the object has no abnormal movement, the corresponding abnormal value
Is 0.
Step S004, accumulating the abnormal degrees of all the frame difference images to obtain comprehensive abnormal degrees, and recognizing that the construction site is abnormal when the comprehensive abnormal degrees are larger than a preset abnormal threshold value.
And accumulating the abnormal degrees of all the frame difference images to obtain the comprehensive abnormal degree, namely accumulating the abnormal degrees from the first frame difference image to the current frame to obtain the comprehensive abnormal degree, and reflecting the real-time abnormal degree on the basis of the reference historical image.
When the comprehensive abnormality degree is greater than a preset abnormality threshold, the abnormality of the construction site is identified, and the monitoring personnel is warned and reminded by sending out measures such as early warning or displaying warning information.
In the embodiment of the invention, the value of the abnormal threshold is 200, and in other embodiments, the value of the abnormal threshold can be adjusted according to the actual situation. And the abnormal threshold at this time is only a reference threshold, and in actual situations, corresponding adjustment is performed according to the construction scene monitored by the specific monitoring picture. For example, in the case of overhead work and ground work, the risk level of the overhead work caused by the abnormality caused by the speed is much higher than that of the ground work, so that the reference threshold value can be lowered in the construction scene of the overhead work, thereby increasing the sensitivity to the risk information in the scene and obtaining the abnormality threshold value in the overhead work.
Further, a monitoring center for displaying monitoring pictures in the construction monitoring system is formed by combining monitoring pictures under a plurality of monitoring visual angles, under the real-time condition, an abnormal degree can be obtained under each monitoring visual angle, the monitoring pictures under all the monitoring visual angles are ordered from large to small, the monitoring pictures are ordered according to the ordering order, and the monitoring picture corresponding to the maximum abnormal degree is in the first position so as to remind monitoring personnel to pay priority attention to the picture.
In summary, the embodiment of the invention acquires the monitoring picture of the construction site, wherein the monitoring picture is a continuous monitoring image; acquiring an information value of each pixel point based on a gray value and a gradient value of the pixel point in each frame of monitoring image; obtaining a corresponding frame difference image by calculating the information value difference of the pixel points at the corresponding positions of every two adjacent frames of monitoring images; identifying objects in the frame difference images, and performing corner matching on two adjacent frame difference images to obtain the moving distance of each object; acquiring a speed abnormal value of a corresponding object according to a time interval and a moving distance between each frame of frame difference image and an adjacent frame of frame difference image, and acquiring an area abnormal value based on the area of each object in the frame difference image; acquiring a normal moving range by utilizing the area change of the frame difference image under a preset period, and acquiring an abnormal adjustment value according to the normal moving range; acquiring the abnormal degree of the corresponding frame difference image through combining the abnormal speed value, the abnormal area value, the abnormal adjustment value and the preset attenuation value of all objects in each frame of frame difference image; and accumulating the abnormal degrees of all the frame difference images to obtain comprehensive abnormal degrees, and identifying that the construction site is abnormal when the comprehensive abnormal degrees are larger than a preset abnormal threshold value. The invention can combine multiple abnormal judgment to obtain accurate abnormal recognition results, reduce false detection of abnormal conditions of a construction site, ensure construction safety and improve construction efficiency.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing description has been directed to specific embodiments of this specification. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and the same or similar parts of each embodiment are referred to each other, and each embodiment mainly describes differences from other embodiments.
The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; the technical solutions described in the foregoing embodiments are modified or some of the technical features are replaced equivalently, so that the essence of the corresponding technical solutions does not deviate from the scope of the technical solutions of the embodiments of the present application, and all the technical solutions are included in the protection scope of the present application.
Claims (8)
1. An anomaly identification method for a construction monitoring system, the method comprising the steps of:
acquiring a monitoring picture of a construction site, wherein the monitoring picture is a continuous monitoring image; acquiring an information value of each pixel point based on a gray value and a gradient value of the pixel point in each frame of monitoring image;
obtaining a corresponding frame difference image by calculating the information value difference of the pixel points at the corresponding positions of every two adjacent frames of monitoring images; identifying objects in the frame difference images, and performing corner matching on two adjacent frame difference images to obtain the moving distance of each object; acquiring a speed abnormal value of a corresponding object according to a time interval and a moving distance between each frame of frame difference image and an adjacent frame of frame difference image, and acquiring an area abnormal value based on the area of each object in the frame difference image;
acquiring a normal moving range by utilizing the area change of a frame difference image under a preset period, and acquiring an abnormal adjustment value according to the normal moving range; acquiring the abnormal degree of the corresponding frame difference image by combining the abnormal speed value and the abnormal area value of all objects in each frame of frame difference image with the abnormal adjustment value and the preset attenuation value;
and accumulating the abnormal degrees of all the frame difference images to obtain comprehensive abnormal degrees, and identifying that the construction site is abnormal when the comprehensive abnormal degrees are larger than a preset abnormal threshold value.
2. The abnormality identification method for a construction monitoring system according to claim 1, wherein the information value acquisition method is:
and respectively normalizing the gray value and the gradient value of each pixel point, and carrying out weighted summation on the obtained two normalization results to obtain the information value of the corresponding pixel point.
3. The abnormality identification method for a construction monitoring system according to claim 1, wherein the frame difference image acquisition method is:
for two adjacent frames of monitoring images, subtracting the information value of the pixel point at the corresponding position in the previous frame of monitoring image from the information value of any pixel point in the next frame of monitoring image, and generating the frame difference image by taking the obtained absolute value of the difference value as the frame difference value at the pixel point and taking the frame difference value as the pixel value of the corresponding pixel point.
4. The abnormality identification method for a construction monitoring system according to claim 1, wherein the moving distance acquisition method is:
and acquiring coordinates of the centroid pixel points of the object in two adjacent frames of frame difference images through the corner matching result, and acquiring the moving distance of the centroid pixel points as the moving distance of the corresponding object according to the coordinates of the centroid pixel points in the two frames of frame difference images.
5. The abnormality identification method for a construction monitoring system according to claim 1, wherein the speed abnormality value acquisition method is:
and obtaining the maximum standard speed of the construction site, taking the ratio of the moving distance of each object to the corresponding time interval as the moving speed of the object, when the moving speed is greater than the maximum standard speed, the object moves abnormally, calculating the speed difference between the moving speed and the maximum standard speed, and taking the ratio of the speed difference in the moving speed as the speed abnormal value.
6. The abnormality identification method for a construction monitoring system according to claim 5, wherein the area abnormality value acquisition method is as follows:
when there is abnormal movement of the object, the area anomaly value is obtained as a result of the area evolution of the object in the frame difference image.
7. The abnormality identification method for a construction monitoring system according to claim 1, wherein the abnormality adjustment value acquisition method is:
and calculating the proportion of the normal moving range to the area of the monitoring image as the abnormal adjustment value.
8. The abnormality identification method for a construction monitoring system according to claim 5, wherein the obtaining method of the degree of abnormality comprises:
and for each object in the frame difference image, when the object moves abnormally, calculating the sum of the speed abnormal value and the area abnormal value, multiplying the abnormal adjustment value to obtain an abnormal value, adjusting the abnormal value by using the preset attenuation value, and summing the abnormal values after the adjustment of all the objects moving abnormally in the frame difference image to obtain the abnormal degree of the corresponding frame difference image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310500967.5A CN116416577B (en) | 2023-05-06 | 2023-05-06 | Abnormality identification method for construction monitoring system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310500967.5A CN116416577B (en) | 2023-05-06 | 2023-05-06 | Abnormality identification method for construction monitoring system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116416577A true CN116416577A (en) | 2023-07-11 |
| CN116416577B CN116416577B (en) | 2023-12-26 |
Family
ID=87054546
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310500967.5A Active CN116416577B (en) | 2023-05-06 | 2023-05-06 | Abnormality identification method for construction monitoring system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116416577B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116703787A (en) * | 2023-08-09 | 2023-09-05 | 中铁建工集团第二建设有限公司 | Building construction safety risk early warning method and system |
| CN117079221A (en) * | 2023-10-13 | 2023-11-17 | 南方电网调峰调频发电有限公司工程建设管理分公司 | Construction safety monitoring method and device for underground engineering of pumping and storing power station |
| CN117197588A (en) * | 2023-11-02 | 2023-12-08 | 南通宝田包装科技有限公司 | Plastic package control early warning method based on temperature identification |
| CN117475381A (en) * | 2023-12-22 | 2024-01-30 | 济宁久邦工程机械设备有限公司 | Real-time monitoring system for operation state of aerial working platform |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090060275A1 (en) * | 2007-08-30 | 2009-03-05 | Casio Computer Co., Ltd. | Moving body image extraction apparatus and computer readable storage medium storing program |
| CN102521842A (en) * | 2011-11-28 | 2012-06-27 | 杭州海康威视数字技术股份有限公司 | Method and device for detecting fast movement |
| CN102622895A (en) * | 2012-03-23 | 2012-08-01 | 长安大学 | Video-based vehicle speed detecting method |
| CN104182990A (en) * | 2014-08-12 | 2014-12-03 | 中国科学院上海微***与信息技术研究所 | A method for acquiring a sequence image motion target area in real-time |
| CN108491762A (en) * | 2018-02-27 | 2018-09-04 | 浙江大华技术股份有限公司 | A kind of detection method and device that human body is fallen |
| CN109101661A (en) * | 2018-09-13 | 2018-12-28 | 东软集团股份有限公司 | The detection method and device of abnormal point in a kind of data sample set |
| CN112304413A (en) * | 2020-09-28 | 2021-02-02 | 梅特勒-托利多(常州)精密仪器有限公司 | Method and device for detecting state of weighing sensor |
| CN114419533A (en) * | 2021-12-09 | 2022-04-29 | 南方海洋科学与工程广东省实验室(湛江) | Deepwater netting damage identification method, device, equipment and storage medium |
| CN114494985A (en) * | 2022-04-18 | 2022-05-13 | 山东西曼克技术有限公司 | Logistics transfer robot work abnormity detection method and system based on image processing |
| CN114842397A (en) * | 2022-05-19 | 2022-08-02 | 华南农业大学 | Real-time old man falling detection method based on anomaly detection |
| CN115082864A (en) * | 2022-07-25 | 2022-09-20 | 青岛亨通建设有限公司 | Building construction safety monitoring system |
| CN115222778A (en) * | 2022-07-29 | 2022-10-21 | 平安科技（深圳）有限公司 | Moving object detection method and device based on optical flow, electronic device and medium |
| CN115240353A (en) * | 2022-09-21 | 2022-10-25 | 江苏海舟安防科技有限公司 | Fire-fighting dangerous case processing method based on Internet of things sensing |
-
2023
- 2023-05-06 CN CN202310500967.5A patent/CN116416577B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090060275A1 (en) * | 2007-08-30 | 2009-03-05 | Casio Computer Co., Ltd. | Moving body image extraction apparatus and computer readable storage medium storing program |
| CN102521842A (en) * | 2011-11-28 | 2012-06-27 | 杭州海康威视数字技术股份有限公司 | Method and device for detecting fast movement |
| CN102622895A (en) * | 2012-03-23 | 2012-08-01 | 长安大学 | Video-based vehicle speed detecting method |
| CN104182990A (en) * | 2014-08-12 | 2014-12-03 | 中国科学院上海微***与信息技术研究所 | A method for acquiring a sequence image motion target area in real-time |
| CN108491762A (en) * | 2018-02-27 | 2018-09-04 | 浙江大华技术股份有限公司 | A kind of detection method and device that human body is fallen |
| CN109101661A (en) * | 2018-09-13 | 2018-12-28 | 东软集团股份有限公司 | The detection method and device of abnormal point in a kind of data sample set |
| CN112304413A (en) * | 2020-09-28 | 2021-02-02 | 梅特勒-托利多(常州)精密仪器有限公司 | Method and device for detecting state of weighing sensor |
| CN114419533A (en) * | 2021-12-09 | 2022-04-29 | 南方海洋科学与工程广东省实验室(湛江) | Deepwater netting damage identification method, device, equipment and storage medium |
| CN114494985A (en) * | 2022-04-18 | 2022-05-13 | 山东西曼克技术有限公司 | Logistics transfer robot work abnormity detection method and system based on image processing |
| CN114842397A (en) * | 2022-05-19 | 2022-08-02 | 华南农业大学 | Real-time old man falling detection method based on anomaly detection |
| CN115082864A (en) * | 2022-07-25 | 2022-09-20 | 青岛亨通建设有限公司 | Building construction safety monitoring system |
| CN115222778A (en) * | 2022-07-29 | 2022-10-21 | 平安科技（深圳）有限公司 | Moving object detection method and device based on optical flow, electronic device and medium |
| CN115240353A (en) * | 2022-09-21 | 2022-10-25 | 江苏海舟安防科技有限公司 | Fire-fighting dangerous case processing method based on Internet of things sensing |
Non-Patent Citations (1)
| Title |
|---|
| 关健, 刘大昕: "基于主成分分析的无监督异常检测", 计算机研究与发展, no. 09, pages 1474 - 1480 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116703787A (en) * | 2023-08-09 | 2023-09-05 | 中铁建工集团第二建设有限公司 | Building construction safety risk early warning method and system |
| CN116703787B (en) * | 2023-08-09 | 2023-10-31 | 中铁建工集团第二建设有限公司 | Building construction safety risk early warning method and system |
| CN117079221A (en) * | 2023-10-13 | 2023-11-17 | 南方电网调峰调频发电有限公司工程建设管理分公司 | Construction safety monitoring method and device for underground engineering of pumping and storing power station |
| CN117079221B (en) * | 2023-10-13 | 2024-01-30 | 南方电网调峰调频发电有限公司工程建设管理分公司 | Construction safety monitoring method and device for underground engineering of pumping and storing power station |
| CN117197588A (en) * | 2023-11-02 | 2023-12-08 | 南通宝田包装科技有限公司 | Plastic package control early warning method based on temperature identification |
| CN117197588B (en) * | 2023-11-02 | 2024-03-05 | 南通宝田包装科技有限公司 | Plastic package control early warning method based on temperature identification |
| CN117475381A (en) * | 2023-12-22 | 2024-01-30 | 济宁久邦工程机械设备有限公司 | Real-time monitoring system for operation state of aerial working platform |
| CN117475381B (en) * | 2023-12-22 | 2024-03-29 | 济宁久邦工程机械设备有限公司 | Real-time monitoring system for operation state of aerial working platform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116416577B (en) | 2023-12-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116416577B (en) | Abnormality identification method for construction monitoring system | |
| CN104123544B (en) | Anomaly detection method and system based on video analysis | |
| CN108802758B (en) | Intelligent security monitoring device, method and system based on laser radar | |
| US20160260306A1 (en) | Method and device for automated early detection of forest fires by means of optical detection of smoke clouds | |
| CN109711318B (en) | Multi-face detection and tracking method based on video stream | |
| US20060170769A1 (en) | Human and object recognition in digital video | |
| CN105894702A (en) | Invasion detecting alarming system based on multi-camera data combination and detecting method thereof | |
| US20160300334A1 (en) | Method and Device for Correction Restoration and Analysis Alarming of Distorted Image | |
| CN107679471A (en) | Indoor occupant sky hilllock detection method based on video monitoring platform | |
| CN103929592A (en) | All-dimensional intelligent monitoring equipment and method | |
| CN105828065B (en) | A kind of video pictures overexposure detection method and device | |
| CN112084963B (en) | Monitoring early warning method, system and storage medium | |
| EP3203447B1 (en) | Detection of fraudulent access at control gates | |
| CN109830078B (en) | Intelligent behavior analysis method and intelligent behavior analysis equipment suitable for narrow space | |
| CN112084813B (en) | Abnormal target detection method, device and storage medium | |
| CN111476964A (en) | Remote forest fire prevention monitoring system and method | |
| KR101454644B1 (en) | Loitering Detection Using a Pedestrian Tracker | |
| CN109034038A (en) | A kind of fire identification device based on multi-feature fusion | |
| CN115953719A (en) | Multi-target recognition computer image processing system | |
| CN111444837A (en) | Temperature measurement method and temperature measurement system for improving face detection availability in extreme environment | |
| JP2000184359A (en) | Monitoring device and system therefor | |
| US20210201074A1 (en) | Method and system for detecting concealed objects using handheld thermal imager | |
| KR20080071770A (en) | System for sensing abnormal phenomenon on realtime and method for controlling the same | |
| CN117241134A (en) | Shooting mode switching method for camera | |
| CN113044694A (en) | Construction site elevator people counting system and method based on deep neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |