CN111582209A - Abnormal behavior supervision method for construction personnel in capital construction site - Google Patents
Abnormal behavior supervision method for construction personnel in capital construction site Download PDFInfo
- Publication number
- CN111582209A CN111582209A CN202010407503.6A CN202010407503A CN111582209A CN 111582209 A CN111582209 A CN 111582209A CN 202010407503 A CN202010407503 A CN 202010407503A CN 111582209 A CN111582209 A CN 111582209A
- Authority
- CN
- China
- Prior art keywords
- gradient
- image
- constructor
- safety helmet
- safety
- 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
- 238000010276 construction Methods 0.000 title claims abstract description 67
- 238000000034 method Methods 0.000 title claims abstract description 26
- 206010000117 Abnormal behaviour Diseases 0.000 title claims abstract description 11
- 238000001514 detection method Methods 0.000 claims abstract description 18
- 230000006399 behavior Effects 0.000 claims abstract description 12
- 238000010606 normalization Methods 0.000 claims description 13
- 239000013598 vector Substances 0.000 claims description 10
- 238000012937 correction Methods 0.000 claims description 4
- 239000000126 substance Substances 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 238000009825 accumulation Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 5
- 238000013473 artificial intelligence Methods 0.000 abstract description 3
- 238000011156 evaluation Methods 0.000 abstract description 2
- 238000012544 monitoring process Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 230000036544 posture Effects 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 238000004080 punching Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/109—Time management, e.g. calendars, reminders, meetings or time accounting
- G06Q10/1091—Recording time for administrative or management purposes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/08—Construction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
- G06Q50/2057—Career enhancement or continuing education service
-
- 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
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
- G06V10/507—Summing image-intensity values; Histogram projection analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- General Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- Economics (AREA)
- Educational Technology (AREA)
- Marketing (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Entrepreneurship & Innovation (AREA)
- Primary Health Care (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention belongs to the field of artificial intelligence technology application, and particularly relates to a method for supervising abnormal behaviors of constructors in a construction site, which comprises the steps of establishing a face database of constructors in the construction site, and storing recent photos of the constructors in the face database; constructing an attendance module of construction site personnel, and checking attendance and checking according to the face information of the construction personnel collected by a monitor or a face data collector; a safety helmet detection module is constructed, and the wearing behavior of the safety helmet of a constructor is judged according to image information collected by the on-site monitor; grading the safety consciousness of each constructor according to the wearing condition of the safety helmet of the constructor in the construction area in working time; according to the invention, attendance evaluation of constructors on the capital construction site is realized through a face recognition technology, and meanwhile, the wearing behavior of the safety helmet is detected through the wearing behavior detection of the safety helmet, so that management personnel can conveniently manage the behaviors of the constructors.
Description
Technical Field
The invention belongs to the field of artificial intelligence technology application, and particularly relates to a method for supervising abnormal behaviors of construction personnel in a capital construction site.
Background
The safety helmet is a key safety guarantee factor for construction personnel in a capital construction site and is one of necessary conditions for the capital construction personnel to enter the capital construction site, but many construction personnel can wear the safety helmet according to various reasons and regulations of capital construction project management, so that safety risk is increased. And because the construction personnel of the capital construction site are numerous, the supervision personnel are limited and can not supervise the monitoring video in real time, which causes the problems of not in place safety supervision and the like. Therefore, the artificial intelligence computer vision technology is utilized to analyze and process the construction site real-time monitoring video, so that the intelligent monitoring of the wearing behavior of the safety helmet of the constructor is realized, the real-time early warning is realized, the safety management efficiency of the supervisors is improved, the occurrence probability of site safety events is reduced, and the orderly development of capital construction projects is ensured.
For the detection of wearing of safety helmets, researchers have conducted a lot of research, mainly including:
(1) the method for identifying the edge of the safety helmet is used for identifying the edge of an object in the head area.
(2) The safety helmet detection algorithm based on the skin color and the Hu moment is used for finding out a face region by using a skin color segmentation method, and after the face region is found, the region above the face is intercepted. And then, extracting the Hu moment of the region image as a feature, then training the extracted Hu moment feature by utilizing an SVM algorithm to obtain a classifier capable of distinguishing a safety helmet from a non-safety helmet, and judging whether the safety helmet is worn or not by using the classifier.
However, the above methods all require a monitoring camera to acquire a frontal area during the work of a worker, so as to acquire facial features and color information of a safety helmet. In the construction process of capital construction, construction workers have different postures and can not ensure that the front face faces the monitoring camera all the time, so that the identification rate is low in practical application and the application effect is poor.
Disclosure of Invention
In order to improve the efficiency of attack detection, the invention provides an intrusion detection method based on a deep residual hash network, which comprises the following steps:
s1, establishing a construction site personnel face database, and storing recent photos of construction personnel into the face database;
s2, constructing an attendance module of the construction site personnel, and checking attendance and checking according to the face information of the construction personnel collected by the monitor or the face data collector;
s3, constructing a safety helmet detection module, and judging the wearing behavior of the safety helmet of the constructor according to the image information collected by the on-site monitor;
and S4, grading the safety consciousness of each constructor according to the wearing condition of the safety helmet of the constructor in the construction area.
Further, constructing the hard hat detection module comprises:
s31, collecting historical data, and dividing the historical data into a positive sample and a negative sample;
s32, inputting HOG characteristics of the positive sample image and the negative sample image into an SVM classifier for classification training;
and S33, extracting the image acquired by the monitor, extracting the HOG characteristic of the image of the monitor, and inputting the characteristic into a trained SVM classifier to obtain a detection result.
Further, the positive sample is a picture of a safety helmet worn, and the negative sample is a picture of a safety helmet not worn.
Further, extracting the image captured by the monitor and extracting the HOG feature of the monitor image includes:
carrying out gamma normalization on the acquired image;
calculating the image gradient after gamma normalization;
dividing the image into cell units, and calculating a gradient histogram of each cell unit; forming a block by several adjacent cell units, and calculating a gradient histogram of each block;
all blocks are concatenated to form the HOG feature of an image.
Further, gamma normalizing the acquired image comprises:
H(x,y)＝H(x,y)γ；
wherein, H (x, y) represents the pixel value of the image pixel (x, y), and γ is the gamma correction coefficient of the pixel value.
Further, calculating the image gradient after gamma normalization comprises:
wherein the content of the first and second substances,
gradient of image at pixel point (x, y), Gx(x, y) is the horizontal gradient of the pixel points of the input image at (x, y), GyAnd (x, y) is the vertical gradient of the pixel points of the input image at (x, y).
Further, calculating the gradient histogram of each cell unit includes:
dividing the image into a cell unit according to 8 multiplied by 8 pixels;
dividing the gradient direction of each cell unit into 9 direction blocks at 0-180 degrees, namely counting the gradient information of 8 multiplied by 8 pixels by using a histogram of 9 intervals;
projecting the image gradient of each pixel in a cell unit to 9 intervals, and accumulating the gradient amplitudes after projection;
and projecting the accumulated gradient amplitude values onto a histogram to obtain a nine-dimensional feature vector of the cell unit.
Further, the gradient histogram of the cell unit is represented as:
wherein, Vk(x, y) is the gradient magnitude accumulation value of all pixels mapped to the k-th interval in the gradient direction; bin (n)kDenotes the kth interval, k ∈ [1, p ]]；Gi(x, y) is the gradient magnitude of the ith pixel value,
the gradient direction of the ith pixel.
Further, calculating the gradient histogram of each block includes:
grouping adjacent 2 x 2 cells into a block;
features in one block are normalized and concatenated to obtain a gradient histogram for each block.
Further, the step of scoring the safety awareness of each constructor according to the wearing condition of the safety helmet of the constructor in the construction area comprises the following steps: collecting the wearing condition of a safety helmet of a constructor in one day, and if the working time of the constructor is [ T ]1,T2]The safety awareness index of the constructor is expressed as:
wherein S isaThe higher the value is, the higher the safety consciousness of the constructor is;
represents the accumulated time T of workers wearing the safety helmet during the working hours1For the worker's work start time, T2The time of the worker going out of work; s1Indicating the proficiency of the worker; s2α + β + 1, α is the weight of the wearing condition of the safety helmet, β is the weight of the proficiency level, and is the weight of the safety learning condition.
According to the invention, attendance evaluation of constructors on the capital construction site is realized through a face recognition technology, and meanwhile, the wearing behavior detection of the safety helmet is realized through the wearing behavior detection of the safety helmet, so that the operation management of constructors is facilitated for managers, and the management efficiency of the managers is improved; meanwhile, the safety cap wearing behavior of the constructors is detected, the construction safety learning condition of the constructors and the skill proficiency of the working skills of the constructors are detected, the safety consciousness assessment algorithm research of the constructors is realized, and references are provided for the capital construction managers for constructors to train safety and improve safety consciousness.
Drawings
FIG. 1 is a flow chart of a method for supervising abnormal behaviors of construction personnel in a capital construction site according to the invention;
FIG. 2 is a schematic diagram of gradient partitioning in the present invention;
FIG. 3 is an illustration of a gradient histogram in accordance with the present invention;
FIG. 4 is a schematic diagram of a relationship between a cell and a block in the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention provides a method for supervising abnormal behaviors of constructors in a capital construction site, which comprises the following steps of:
s1, establishing a construction site personnel face database, and storing recent photos of construction personnel into the face database;
s2, constructing an attendance module of the construction site personnel, and checking attendance and checking according to the face information of the construction personnel collected by the monitor or the face data collector;
s3, constructing a safety helmet detection module, and judging the wearing behavior of the safety helmet of the constructor according to the image information collected by the on-site monitor;
and S4, grading the safety consciousness of each constructor according to the wearing condition of the safety helmet of the constructor in the construction area.
In this embodiment, when the face database of the constructor is established, the recent photos of the constructor are stored in the face database.
The attendance checking module in the embodiment is also an important part for detecting the abnormal conditions of the construction site, and can judge whether the worker has the phenomenon of late arrival and early retreat according to the time of punching the card. For example, the work time of the construction site is 9 am, 8 hours of work, and 5 pm off work, and if the time for the worker to punch the card is earlier than 9 am, the worker can be considered as normal attendance; if the attendance of the worker is recorded as late arrival before 11 noon after 9 o' clock; if the card is punched after 11 o' clock, the worker is considered to be spacious in work for half a day.
The invention is most important to detect the wearing condition of the safety helmet of personnel in the capital construction field. Firstly, a detection mechanism needs to be constructed, the embodiment crawls pictures from a network, and divides the crawled pictures into positive samples and negative samples, wherein the positive samples are pictures of staff who normally wear a safety helmet and comprise positive photos, side photos and photos during construction, and the negative samples are photos of people who do not wear the safety helmet and do not correctly wear the safety helmet and also comprise the positive photos. Side photographs and photographs being constructed;
extracting HOG characteristics of the obtained samples, inputting the characteristics into an SVM classifier for training, and obtaining the trained SVM classifier; .
Before detection, a monitoring camera needs to be installed on a capital construction site, particularly in a high-risk operation area, and pictures of construction personnel during working are collected;
and extracting the HOG characteristics of the acquired image, and inputting the extracted characteristics into a trained SVM classifier to obtain whether the wearing of the artificial safety helmet in the image is standard or not.
The method for extracting the image HOG features comprises the following steps:
carrying out gamma normalization on the collected image
In order to reduce the influence of local shadow of an image caused by illumination factors, firstly, a Gamma correction method is adopted to standardize the color space of an input image. The normalization process is represented as:
H(x,y)＝H(x,y)γ；
h (x, y) represents a pixel value of the image pixel (x, y), γ is a gamma correction coefficient of the pixel value, and a preferred value of γ is 0.5.
(II) calculating image gradient after gamma normalization
The local characteristics of the image such as gray scale, color, texture mutation and the like form the edge of the image object, so that adjacent pixel points in the image have less change, the regional change is not obvious, the gradient amplitude is smaller, and adjacent pixel points at the edge part have larger change, and the gradient amplitude is larger. The gradient in the image is represented as the first derivative of the image function, but the gradient information of the image can be calculated more simply, quickly and effectively by using a one-dimensional discrete differential template, and the horizontal direction gradient and the vertical direction gradient calculation formula are as follows:
Gx(x,y)＝H(x+1,y)-H(x-1,y)；
Gy(x,y)＝H(x,y+1)-H(x,y-1)；
wherein G isx(x, y) is the horizontal gradient of the pixel points of the input image at (x, y), GyAnd (x, y) is the vertical gradient of the pixel points of the input image at (x, y). The gradient amplitude calculation formula of the pixel point at (x, y) is as follows:
the gradient direction calculation formula of the pixel point in (x, y) is as follows:
wherein the content of the first and second substances,
is the gradient of the image at the pixel point (x, y).
(III) construction of gradient histograms of cell units
The image is divided into small cell cells, each cell comprising 8 x 8 pixels, and a gradient histogram is calculated for each cell. As shown in fig. 2, the gradient direction of each cell is divided into 9 direction blocks at 0-180 °, that is, the gradient information of the 8 × 8 pixels is counted by using a histogram of 9 bins (bins), so that each cell corresponds to a 9-dimensional feature vector.
For each pixel in the cell, the gradient direction of the pixel (x, y) is calculated, as shown in fig. 2, the gradient direction of the pixel is projected in 9 bins of the gradient direction division diagrams Z1-Z9, then the gradient amplitudes of the pixels mapped to each bin are accumulated, finally the gradient amplitude accumulated value of each bin is mapped to a histogram, the gradient direction histogram of the cell can be obtained, and the gradient amplitude accumulated value of the 9 bins forms the 9-dimensional feature vector of the cell. For example, if the gradient direction calculated by a certain pixel in a cell is 20 to 40 °, the pixel is mapped to the 2 nd bin, the gradient magnitude of the pixel is accumulated, and the pixel is mapped to the histogram corresponding to the 2 nd bin.
The histogram of the gradient of the cell unit is shown in fig. 3, in which the amplitude value of the k-th interval is expressed as:
wherein, Vk(x, y) is the gradient magnitude accumulation value of all pixels whose gradient direction is mapped to the k-th interval, Vk(x, y) the histogram is mapped to the gradient histogram of the cell unit; bin (n)kDenotes the kth interval, k ∈ [1, p ]]；Gi(x, y) is the gradient magnitude of the ith pixel value,
the gradient direction of the ith pixel.
(IV) constructing a gradient histogram for each block
Several adjacent cell units are grouped into a block (block), and usually adjacent 2 × 2 cell units are grouped into a block, and a schematic diagram of a cell and a block is shown in fig. 4. The gradient direction histogram of a block is formed by connecting feature vectors of four cell units (cells) in the block in series, and the Hog feature of a block is 4 × 9-36 dimensions according to the fact that a cell unit is a 9-dimensional Hog feature.
After the histogram of the block is constructed, local contrast normalization needs to be performed on the gradient, so that the problems that the change range of the gradient strength is too large and the like caused by local illumination, foreground and background contrast change and the like are solved.
The method comprises the following specific steps:
assume that the feature vector for a cell is: x ═ x1,x2,...,x9Then the normalization formula of this vector is as follows:
firstly, obtaining the square sum of each element of the vector, and performing evolution on the square sum; then dividing each element of the vector by the data after the evolution to finish the normalization; wherein the content of the first and second substances,
a characterization representation of the ith cell is shown.
Then for a block consisting of 2 × 2, only the feature vectors of each cell need to be concatenated, and the features of a block are represented as:
(V) HOG feature extraction
And (4) connecting the feature description substrings of all blocks in the image to obtain the HOG feature of the image.
After passing through above detection module, according to the safety helmet condition of wearing at the construction area of constructor operating time to grade every constructor's safety consciousness and include: collecting the wearing condition of a safety helmet of a constructor in one day, and if the working time of the constructor is [ T ]1,T2]The safety awareness index of the constructor is expressed as:
wherein S isaThe higher the value is, the higher the safety consciousness of the constructor is;
represents the accumulated time T of workers wearing the safety helmet during the working hours1For the worker's work start time, T2The time of the worker going out of work; s1Indicating the proficiency of the worker; s2Indicating worker's safe learning situationα + β + 1, α is the weight of the helmet wearing condition, β is the weight of the skill level, and is the weight of the safety learning condition.
As an alternative embodiment, the worker consciousness index and attendance can be combined to serve as an index of abnormal conditions of the capital construction site. The proficiency of the worker can be hooked with the related skill license of the worker in the specific implementation process, or primary proficiency workers and secondary proficiency workers are divided according to the age and the working age of the worker, and the like, wherein the s1 value of the high-level proficiency workers is higher than that of the low-level proficiency workers; for the safety learning condition of a worker, the embodiment mainly uses two parameters, namely safety learning and/or safety examination, the safety learning is the proportion of the number of times of the worker for taking part in the safety learning each month to the total number of times of the worker for taking part in the month, and the safety examination takes the safety examination score of the current quarter as the calculation parameter of the safety consciousness index of the next quarter.
Considering that wearing behaviors of construction personnel on a capital construction site safety helmet are mainly taken as main parameters to be considered, the value of the weight alpha is generally more than or equal to 0.5; the proficiency of workers is also related to the probability of accidents on the site, and according to incomplete statistics, the probability of accidents of more proficient workers is smaller, so that the weight beta is generally less than or equal to 0.3-0.4; the safe learning condition has a certain preventive effect, and the weight value is generally 0.2-0.1.
In the description of the present invention, it is to be understood that the terms "coaxial", "bottom", "one end", "top", "middle", "other end", "upper", "one side", "top", "inner", "outer", "front", "center", "both ends", and the like, indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience of description and simplicity of description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, are not to be construed as limiting the present invention.
In the present invention, unless otherwise expressly stated or limited, the terms "mounted," "disposed," "connected," "fixed," "rotated," and the like are to be construed broadly, e.g., as meaning fixedly connected, detachably connected, or integrally formed; can be mechanically or electrically connected; the terms may be directly connected or indirectly connected through an intermediate, and may be communication between two elements or interaction relationship between two elements, unless otherwise specifically limited, and the specific meaning of the terms in the present invention will be understood by those skilled in the art according to specific situations.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (10)
1. A method for supervising abnormal behaviors of constructors on a capital construction site is characterized by comprising the following steps:
s1, establishing a construction site personnel face database, and storing recent photos of construction personnel into the face database;
s2, constructing an attendance module of the construction site personnel, and checking attendance and checking according to the face information of the construction personnel collected by the monitor or the face data collector;
s3, constructing a safety helmet detection module, and judging the wearing behavior of the safety helmet of the constructor according to the image information collected by the on-site monitor;
and S4, grading the safety consciousness of each constructor according to the wearing condition of the safety helmet of the constructor in the construction area.
2. The method for supervising abnormal behaviors of construction personnel on a capital construction site according to claim 1, wherein the step of constructing the safety helmet detection module comprises the following steps:
s31, collecting historical data, and dividing the historical data into a positive sample and a negative sample;
s32, inputting HOG characteristics of the positive sample image and the negative sample image into an SVM classifier for classification training;
and S33, extracting the image acquired by the monitor, extracting the HOG characteristic of the image of the monitor, and inputting the characteristic into a trained SVM classifier to obtain a detection result.
3. The method as claimed in claim 2, wherein the positive sample is a picture of a helmet, and the negative sample is a picture of a helmet.
4. The method as claimed in claim 2, wherein the extracting the image collected by the monitor and the HOG feature of the image of the monitor comprises:
carrying out gamma normalization on the acquired image;
calculating the image gradient after gamma normalization;
dividing the image into cell units, and calculating a gradient histogram of each cell unit;
forming a block by several adjacent cell units, and calculating a gradient histogram of each block;
all blocks are concatenated to form the HOG feature of an image.
5. The method for supervising abnormal behaviors of construction personnel on a capital construction site according to claim 4, wherein the step of carrying out gamma normalization on the acquired images comprises the following steps:
H(x,y)＝H(x,y)γ；
wherein, H (x, y) represents the pixel value of the image pixel (x, y), and γ is the gamma correction coefficient of the pixel value.
6. The method as claimed in claim 4, wherein the step of calculating the gradient of the image after gamma normalization comprises:
wherein the content of the first and second substances,
gradient of image at pixel point (x, y), Gx(x, y) is the horizontal gradient of the pixel points of the input image at (x, y), GyAnd (x, y) is the vertical gradient of the pixel points of the input image at (x, y).
7. The method as claimed in claim 4, wherein the calculating of the histogram of gradient of each cell unit comprises:
dividing the image into a cell unit according to 8 multiplied by 8 pixels;
dividing the gradient direction of each cell unit into 9 direction blocks at 0-180 degrees, namely counting the gradient information of 8 multiplied by 8 pixels by using a histogram of 9 intervals;
projecting the image gradient of each pixel in a cell unit to 9 intervals, and accumulating the gradient amplitudes after projection;
and projecting the accumulated gradient amplitude values onto a histogram to obtain a nine-dimensional feature vector of the cell unit.
8. The method for supervising abnormal behaviors of construction personnel on a capital construction site according to claim 7, wherein the gradient histogram of the cell unit is represented as follows:
wherein, Vk(x, y) is the gradient magnitude accumulation value of all pixels mapped to the k-th interval in the gradient direction; bin (n)kDenotes the kth interval, k ∈ [1, p ]]；Gi(x, y) is the gradient magnitude of the ith pixel value,
the gradient direction of the ith pixel.
9. The method as claimed in claim 4, wherein the calculating the histogram of gradient of each block includes:
grouping adjacent 2 x 2 cells into a block;
features in one block are normalized and concatenated to obtain a gradient histogram for each block.
10. The method for supervising abnormal behaviors of constructors on a capital construction site according to claim 1, wherein the step of scoring the safety awareness of each constructor according to the wearing condition of the safety helmet of the constructor in a construction area during working time comprises the following steps: collecting the wearing condition of a safety helmet of a constructor in one day, and if the working time of the constructor is [ T ]1,T2]The safety awareness index of the constructor is expressed as:
wherein S isaThe higher the value is, the higher the safety consciousness of the constructor is;
represents the accumulated time T of workers wearing the safety helmet during the working hours1For the worker's work start time, T2The time of the worker going out of work; s1Indicating the proficiency of the worker; s2α + β + 1, α is the weight of the wearing condition of the safety helmet, β is the weight of the proficiency level, and is the weight of the safety learning condition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010407503.6A CN111582209A (en) | 2020-05-14 | 2020-05-14 | Abnormal behavior supervision method for construction personnel in capital construction site |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010407503.6A CN111582209A (en) | 2020-05-14 | 2020-05-14 | Abnormal behavior supervision method for construction personnel in capital construction site |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111582209A true CN111582209A (en) | 2020-08-25 |
Family
ID=72117397
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010407503.6A Pending CN111582209A (en) | 2020-05-14 | 2020-05-14 | Abnormal behavior supervision method for construction personnel in capital construction site |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111582209A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112037406A (en) * | 2020-08-27 | 2020-12-04 | 江门明浩电力工程监理有限公司 | Intelligent construction site access control method, system and equipment |
| CN113128876A (en) * | 2021-04-22 | 2021-07-16 | 北京房江湖科技有限公司 | Image-based object management method, device and computer-readable storage medium |
| CN113903068A (en) * | 2021-10-19 | 2022-01-07 | 深圳市中博科创信息技术有限公司 | Stranger monitoring method, device and equipment based on human face features and storage medium |
| CN114386939A (en) * | 2021-12-27 | 2022-04-22 | 浙江新导向标识有限公司 | Construction supervision method and system, storage medium and intelligent terminal |
| CN114445925A (en) * | 2022-04-11 | 2022-05-06 | 深圳市润璟元信息科技有限公司 | Facial recognition intelligent attendance system capable of being automatically loaded and deleted |
| CN116934091A (en) * | 2023-07-25 | 2023-10-24 | 广东至衡工程管理有限公司 | Construction site monitoring method and device, computer equipment and storage medium |
| CN117333929A (en) * | 2023-12-01 | 2024-01-02 | 贵州省公路建设养护集团有限公司 | Method and system for identifying abnormal personnel under road construction based on deep learning |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103745226A (en) * | 2013-12-31 | 2014-04-23 | 国家电网公司 | Dressing safety detection method for worker on working site of electric power facility |
| CN104063722A (en) * | 2014-07-15 | 2014-09-24 | 国家电网公司 | Safety helmet identification method integrating HOG human body target detection and SVM classifier |
| US20160191995A1 (en) * | 2011-09-30 | 2016-06-30 | Affectiva, Inc. | Image analysis for attendance query evaluation |
| CN106778609A (en) * | 2016-12-15 | 2017-05-31 | 国网浙江省电力公司杭州供电公司 | A kind of electric power construction field personnel uniform wears recognition methods |
| CN108268887A (en) * | 2017-12-15 | 2018-07-10 | 西安电子科技大学 | Driver's awareness of safety appraisal procedure based on virtual driving and EEG detections |
| CN108597050A (en) * | 2018-04-20 | 2018-09-28 | 重庆睿豪科技发展有限公司 | Construction site based on Internet of Things manages system |
| CN108765610A (en) * | 2018-05-14 | 2018-11-06 | 北京洛斯达数字遥感技术有限公司 | Attendance management method for work transmission line scene |
-
2020
- 2020-05-14 CN CN202010407503.6A patent/CN111582209A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160191995A1 (en) * | 2011-09-30 | 2016-06-30 | Affectiva, Inc. | Image analysis for attendance query evaluation |
| CN103745226A (en) * | 2013-12-31 | 2014-04-23 | 国家电网公司 | Dressing safety detection method for worker on working site of electric power facility |
| CN104063722A (en) * | 2014-07-15 | 2014-09-24 | 国家电网公司 | Safety helmet identification method integrating HOG human body target detection and SVM classifier |
| CN106778609A (en) * | 2016-12-15 | 2017-05-31 | 国网浙江省电力公司杭州供电公司 | A kind of electric power construction field personnel uniform wears recognition methods |
| CN108268887A (en) * | 2017-12-15 | 2018-07-10 | 西安电子科技大学 | Driver's awareness of safety appraisal procedure based on virtual driving and EEG detections |
| CN108597050A (en) * | 2018-04-20 | 2018-09-28 | 重庆睿豪科技发展有限公司 | Construction site based on Internet of Things manages system |
| CN108765610A (en) * | 2018-05-14 | 2018-11-06 | 北京洛斯达数字遥感技术有限公司 | Attendance management method for work transmission line scene |
Non-Patent Citations (2)
| Title |
|---|
| JIE LI等: "Safety helmet wearing detection based on image processing and machine learning", 《2017 NINTH INTERNATIONAL CONFERENCE ON ADVANCED COMPUTATIONAL INTELLIGENCE (ICACI)》 * |
| 张宏等: "结合智能安全帽的建筑工人施工安全行为绩效考核及激励机制", 《中国安全生产科学技术》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112037406A (en) * | 2020-08-27 | 2020-12-04 | 江门明浩电力工程监理有限公司 | Intelligent construction site access control method, system and equipment |
| CN113128876A (en) * | 2021-04-22 | 2021-07-16 | 北京房江湖科技有限公司 | Image-based object management method, device and computer-readable storage medium |
| CN113903068A (en) * | 2021-10-19 | 2022-01-07 | 深圳市中博科创信息技术有限公司 | Stranger monitoring method, device and equipment based on human face features and storage medium |
| CN114386939A (en) * | 2021-12-27 | 2022-04-22 | 浙江新导向标识有限公司 | Construction supervision method and system, storage medium and intelligent terminal |
| CN114386939B (en) * | 2021-12-27 | 2022-11-25 | 浙江新导向标识有限公司 | Construction supervision method and system, storage medium and intelligent terminal |
| CN114445925A (en) * | 2022-04-11 | 2022-05-06 | 深圳市润璟元信息科技有限公司 | Facial recognition intelligent attendance system capable of being automatically loaded and deleted |
| CN114445925B (en) * | 2022-04-11 | 2022-07-22 | 深圳市润璟元信息科技有限公司 | Facial recognition intelligent attendance system capable of being automatically loaded and deleted |
| CN116934091A (en) * | 2023-07-25 | 2023-10-24 | 广东至衡工程管理有限公司 | Construction site monitoring method and device, computer equipment and storage medium |
| CN117333929A (en) * | 2023-12-01 | 2024-01-02 | 贵州省公路建设养护集团有限公司 | Method and system for identifying abnormal personnel under road construction based on deep learning |
| CN117333929B (en) * | 2023-12-01 | 2024-02-09 | 贵州省公路建设养护集团有限公司 | Method and system for identifying abnormal personnel under road construction based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111582209A (en) | Abnormal behavior supervision method for construction personnel in capital construction site | |
| CN110826538B (en) | Abnormal off-duty identification system for electric power business hall | |
| CN111598040B (en) | Construction worker identity recognition and safety helmet wearing detection method and system | |
| CN111967393B (en) | Safety helmet wearing detection method based on improved YOLOv4 | |
| CN105160297B (en) | Masked man's event automatic detection method based on features of skin colors | |
| CN110348312A (en) | A kind of area video human action behavior real-time identification method | |
| CN105469105A (en) | Cigarette smoke detection method based on video monitoring | |
| CN113516076A (en) | Improved lightweight YOLO v4 safety protection detection method based on attention mechanism | |
| CN110189447B (en) | Intelligent community gate control system based on face identity recognition | |
| CN113411542A (en) | Intelligent working condition monitoring equipment | |
| CN106056079A (en) | Image acquisition device and facial feature occlusion detection method | |
| WO2022262257A1 (en) | Method for recognizing sleep behavior of operator on duty in field of inspection | |
| CN111126219A (en) | Transformer substation personnel identity recognition system and method based on artificial intelligence | |
| CN112183438B (en) | Image identification method for illegal behaviors based on small sample learning neural network | |
| CN112163572A (en) | Method and device for identifying object | |
| CN111667655A (en) | Infrared image-based high-speed railway safety area intrusion alarm device and method | |
| CN112183472A (en) | Method for detecting whether test field personnel wear work clothes or not based on improved RetinaNet | |
| CN114049589A (en) | Transformer substation monitoring system based on artificial intelligence | |
| CN111476083A (en) | Automatic identification method for wearing of safety helmet of electric power staff | |
| CN109001210B (en) | System and method for detecting aging and cracking of sealing rubber strip of civil air defense door | |
| CN112613449A (en) | Safety helmet wearing detection and identification method and system based on video face image | |
| CN113298077A (en) | Transformer substation foreign matter identification and positioning method and device based on deep learning | |
| CN106372566A (en) | Digital signage-based emergency evacuation system and method | |
| CN110543809A (en) | Electric power operation site risk supervision intrusion video identification method | |
| CN114037873A (en) | Transformer substation secondary circuit protection pressing plate detection system and method based on artificial intelligence |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200825 |
|
| RJ01 | Rejection of invention patent application after publication |