CN111598040A - Construction worker identity identification and safety helmet wearing detection method and system - Google Patents
Construction worker identity identification and safety helmet wearing detection method and system Download PDFInfo
- Publication number
- CN111598040A CN111598040A CN202010447288.2A CN202010447288A CN111598040A CN 111598040 A CN111598040 A CN 111598040A CN 202010447288 A CN202010447288 A CN 202010447288A CN 111598040 A CN111598040 A CN 111598040A
- Authority
- CN
- China
- Prior art keywords
- construction
- safety helmet
- data set
- neural network
- yolo
- 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
- 238000010276 construction Methods 0.000 title claims abstract description 93
- 238000001514 detection method Methods 0.000 title claims abstract description 46
- 238000012549 training Methods 0.000 claims abstract description 31
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 25
- 238000013528 artificial neural network Methods 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 18
- 238000012360 testing method Methods 0.000 claims abstract description 14
- 238000003062 neural network model Methods 0.000 claims abstract description 13
- 238000012795 verification Methods 0.000 claims abstract description 12
- 238000010606 normalization Methods 0.000 claims abstract description 11
- 238000013135 deep learning Methods 0.000 claims abstract description 6
- 238000000034 method Methods 0.000 claims description 26
- 230000001815 facial effect Effects 0.000 claims description 10
- 230000006870 function Effects 0.000 claims description 10
- 238000013527 convolutional neural network Methods 0.000 claims description 9
- 238000005286 illumination Methods 0.000 claims description 5
- 230000008859 change Effects 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 238000013480 data collection Methods 0.000 claims description 3
- 230000007423 decrease Effects 0.000 claims description 2
- 230000002238 attenuated effect Effects 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 10
- 238000004590 computer program Methods 0.000 description 7
- 238000007689 inspection Methods 0.000 description 6
- 230000006399 behavior Effects 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 208000019462 Occupational injury Diseases 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011056 performance test Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- 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
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/55—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C9/00—Individual registration on entry or exit
- G07C9/30—Individual registration on entry or exit not involving the use of a pass
- G07C9/32—Individual registration on entry or exit not involving the use of a pass in combination with an identity check
- G07C9/37—Individual registration on entry or exit not involving the use of a pass in combination with an identity check using biometric data, e.g. fingerprints, iris scans or voice recognition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Human Computer Interaction (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Multimedia (AREA)
- Databases & Information Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a construction worker identity identification and safety helmet wearing detection method and a system, comprising the following steps: collecting identity information of construction workers and worker images monitored by a real-name system entrance; adding a label to the image in the public data set SHWD, and performing normalization processing; building a system architecture based on a YOLO V3 and Light cnn-29 neural network algorithm; training YOLO v3 on a public data set SHWD, inputting marked images, and detecting safety helmets and face parts; training Light CNN-29; testing the recognition precision and recall rate under various construction environments, and correcting; and (4) carrying out identity verification on personnel entering the construction site by using the corrected neural network model, and carrying out safety helmet detection on the personnel entering the construction site. According to the invention, by means of a deep learning algorithm, the identity of personnel entering the construction site is verified, and the safety helmet detection is carried out on the personnel entering the construction site, so that personnel who are not the same project and who do not wear the safety helmet are prevented from entering the construction site, and the safety management capability of the construction site is effectively improved.
Description
Technical Field
The invention relates to the technical field of computer AI, in particular to a construction worker identity identification and safety helmet wearing detection method.
Background
For decades, security has been a concern for the construction industry. Occupational injuries in the construction field are mainly derived from unsafe behavior, unsafe environments and unsafe mechanical tools, where unsafe environments and behaviors are considered to be the main causes of construction accidents. At present, the inspection of unsafe behaviors and environments in the building field mainly depends on the manpower detection of a construction site, and the manpower resources which can be allocated to the safety detection are limited due to the heavy construction engineering task. With the development of computers and information technologies, more and more intelligent monitoring and detection technologies are applied to security detection in the building field.
At the construction site, the safety behavior of workers and the inspection of safety protection settings often depend on the level of patrol by safety officers. The identity of personnel entering a construction site and the safety helmet inspection are mainly determined by on-site manual inspection, the mobility of construction workers is high, the identity of each personnel entering the construction site cannot be determined by the manual inspection, and the safety helmet inspection of the construction workers is usually neglected.
Disclosure of Invention
According to the method for identifying the identity of the construction worker and detecting the wearing of the safety helmet, computer vision is introduced into construction safety detection, and manual intelligent technology is used for replacing manual detection in the traditional construction safety.
In order to achieve the purpose, the invention adopts the following technical scheme:
a construction worker identity identification and safety helmet wearing detection method comprises the following steps:
the method comprises the following steps:
downloading a public data set SHWD, and collecting identity information of construction workers and worker images monitored by a real-name system entrance;
adding a label to the image in the public data set SHWD, naming and storing the collected image by worker identity information, and performing normalization processing;
building a system architecture based on a YOLO V3 and Light cnn-29 neural network algorithm;
training YOLO v3 on the common data set SHWD and inputting the labeled SHWD data set image to YOLO v 3;
taking the image detected as a face area by YOLO v3 as a face recognition training set, and training Light CNN-29;
respectively detecting the performances of a YOLO v3 neural network system and a Light cnn-29 neural network system, testing the recognition accuracy and the recall rate under various construction environments, and correcting;
further, downloading a public data set SHWD, and collecting identity information of construction workers and images of workers monitored by a real-name system entrance; the method comprises the following steps:
first, the department of project manager and each team builder are collected, the face under the condition of wearing the safety helmet is self-photographed, and the photograph information is recorded as follows: putting the name, team and age into a database;
secondly, intercepting facial image information of construction workers through a camera installed at an entrance of a real-name channel of a construction site, expanding a data set, and classifying and storing the data set according to the actual construction site condition;
finally, a common data set SHWD is downloaded, which is the common data set used for headgear wear detection.
Further, tagging images in the public data set SHWD, saving the collected images in the name of the worker identity information, and performing normalization processing, including:
normalizing the collected image to 416 × 416 size, wherein the image content is maximized and comprises human faces and safety helmets;
further comprising:
labeling the classified samples and the data samples labeled with the position boundaries; for the collected data set, LabelImg was used to label the worker's facial features and helmet features, and these annotations were saved as XML files in paschalloc format for Python reading.
Further, the introducing Haar cascade classifier comprises:
the Haar cascade classifier locates the position of the face through an expandable window, and takes the highest score from a prediction window to locate the face.
On the other hand, the invention also discloses a construction worker identity identification and safety helmet wearing detection system, which comprises the following units:
the data collection unit is used for respectively collecting identity information of construction workers and corresponding image information of the safety helmet wearing;
the data processing unit is used for adding labels to the collected image information, storing the labels and performing normalization processing;
the model training unit is used for inputting the image added with the label into a convolutional neural network and training a neural network model;
the model correction unit is used for detecting the performance of the trained convolutional neural network of the trained neural network model, testing the recognition precision and recall rate in various environments and correcting;
and the data detection unit is used for carrying out identity verification on personnel entering the construction site and carrying out safety helmet detection on the personnel entering the construction site by utilizing the corrected neural network model.
According to the technical scheme, the method and the system for identifying the identity of the construction worker and detecting the wearing of the safety helmet have the following beneficial effects:
in summary, according to the construction worker identity recognition and safety helmet wearing detection method based on deep learning, the facial image of the construction worker is collected, a supervised learning mode is adopted, the Python programming language and the YOLO v3 network framework are used for extracting the facial features and the safety helmet features, the image information is learned and tested, and the identity verification of the worker and the safety helmet wearing detection are achieved. Through a large amount of data acquisition and training, the accuracy rate of worker identity verification and the detection and identification rate of the safety helmet are both up to over 90%.
Compared with the traditional face recognition method, the method can recognize the face of a construction worker in various weather and illumination environments, can achieve more than 85% of recognition paths under the condition of shielding, and has strong robustness; compared with a general safety helmet detection algorithm, the method can expand the function of worker identity verification on the basis of safety helmet identification, and can realize real-time monitoring.
Drawings
FIG. 1 is a flow chart of a method of the present invention;
FIG. 2 is a schematic diagram of the method of the present invention;
FIG. 3 is a diagram of the YOLO v3 network architecture of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention.
As shown in fig. 1, the method for identifying identity of construction worker and detecting wearing of safety helmet in this embodiment includes:
the method comprises the following steps:
downloading a public data set SHWD, and collecting identity information of construction workers and worker images monitored by a real-name system entrance;
adding a label to the image in the public data set SHWD, naming and storing the collected image by worker identity information, and performing normalization processing;
building a system architecture based on a YOLO V3 and Light cnn-29 neural network algorithm;
training YOLO v3 on the common data set SHWD and inputting the labeled SHWD data set image to YOLO v 3;
taking the image detected as a face area by YOLO v3 as a face recognition training set, and training Light CNN-29;
respectively detecting the performances of a YOLO v3 neural network system and a Light cnn-29 neural network system, testing the recognition accuracy and the recall rate under various construction environments, and correcting;
and verifying the identity of the personnel entering the construction site by using the corrected neural network model, and detecting the safety helmet of the personnel entering the construction site.
The following is a detailed description:
safety management applied to a construction site, as shown in fig. 2, comprises the following steps:
step 1: downloading a public data set SHWD, and collecting identity information of construction workers and worker images monitored by a real-name system entrance;
the existing public data set does not support the training requirement of the invention, and a worker face information and safety helmet library is built by the user. First, we collected the department of project manager and each team builder, self-photographed the face with the crash helmet worn, and the photo information as follows: the name + team + age (e.g., the lisk builder 31) is placed in the database. Secondly, in order to guarantee the generalization capability of the convolutional neural network, human face image information of construction workers is intercepted through a camera installed at an entrance of a real-name channel of a construction site, the image is intercepted according to the standard that human face features can be clearly obtained and a safety helmet is worn, a data set is enlarged, and the data set is classified and stored according to actual construction site conditions.
The method specifically comprises the following steps:
first, the department of project manager and each team builder are collected, the face under the condition of wearing the safety helmet is self-photographed, and the photograph information is recorded as follows: putting the name, team and age into a database;
secondly, intercepting facial image information of construction workers through a camera installed at an entrance of a real-name channel of a construction site, expanding a data set, and classifying and storing the data set according to the actual construction site condition;
finally, the common data set SHWD is downloaded. The SHWD is a common data set for headgear wear detection, comprising 7581 images of 9044 human face-worn headgear objects (positive) and 111514 normal head objects (unworn or negative).
Step 2: adding a label to the image in the public data set SHWD, naming and storing the collected image by worker identity information, and performing normalization processing;
the input image size of the YOLO v3 neural network is 416 × 416, we normalize the collected image to 416 × 416 size, and the image content is maximized to include faces and helmets. Labeling of classified samples and location-bound labeled data samples is required. For the collected data set, LabelImg was used to label the worker's facial features and helmet features, and these annotations were saved as XML files in PASCAL VOC format for Python reading.
And step 3: building a system architecture based on a YOLO V3 and Light cnn-29 neural network algorithm;
the neural network is developed by adopting OpenCV and CAFFE frameworks based on python 3.7 programming language of a Windows system. Firstly, a system Python environment is built, an image description function is defined, then trained YOLO v3 and Light cnn-29 algorithms are introduced, a Haar cascade classifier is introduced, and finally a Utils module is defined, wherein the Utils module can read results of recognition of YOLO v3 and Light cnn-29 and output in a combined mode.
The Haar cascade classifier locates the position of the face through an expandable window, and takes the highest score from the prediction window to locate the face.
And 4, step 4: training YOLO v3 on the common data set SHWD and inputting the labeled SHWD data set image to YOLO v 3;
the neural network was trained on device i7-8700k, NVIDIA GTX 10606 GB.
Creating a VOC format training set, which mainly comprises:
the method comprises the steps of firstly creating an indications folder which is mainly used for storing image marks in an xml format, then creating an ImageSets folder which stores test.txt, train.txt and val.txt and is used for determining training images and verification images in the SHWD data set, and finally creating a JPEGImages folder which is used for storing all images in the SHWD data set.
And 5: taking the image detected as a face area by YOLO v3 as a face recognition training set, and training LightCNN-29;
the method mainly comprises the following steps:
the Light cnn model was implemented using the open source deep learning framework Caffe Dropout for fully connected layers with the scale set to 0.7. for convolutional layers and fully connected layers except for fc2 layer, the momentum is set to 0.9 and the weight attenuation is set to 5 × 104As the identity in the training dataset increases, fc2 contains a large number of parameters that are not used for feature extraction to avoid overfitting, we add weights to the fc2 layer, attenuating it to 5 × 103Initial learning rate set to 1 × 103Then gradually decreases to 5 × 105。
Step 6: respectively detecting the performances of a YOLO v3 neural network system and a Light cnn-29 neural network system, testing the recognition accuracy and the recall rate under various construction environments, and correcting;
when evaluating a target algorithm identification algorithm, two indexes of accuracy and recall rate are usually used for measuring the accuracy of the algorithm, and a true case TP, a false positive case FP and a false negative case FN are defined at first; TP represents the number of the construction workers without the safety helmet after the algorithm is operated, FP represents the number of the construction workers without the safety helmet, but the result is not accurate. FN is the number of misjudged construction workers who did not wear safety helmets. The target recognition accuracy represents the proportion of true case samples TP to total samples (TP + FP), and is used to measure the reliability of the recognition performance. The recall rate indicates the proportion of the true positive sample TP to the total positive sample (TP + FN), which are commonly used evaluation indexes for target recognition.
The specific calculation formula is as follows:
the time of target identification is also an important performance of the measurement algorithm, and here, the identification time is also taken as a standard of the detection algorithm;
wherein the detection of the accuracy of the test identification under various environments comprises:
different weather environment and the different times on same day, illumination all can change, can all exert an influence to the definition of image, and the change of number of people and facial sheltering from all can exert an influence to the recognition accuracy of algorithm, take these two items into account, and concrete classification is as shown in table 1:
TABLE 1 data classification Table
The expression of face recognition is mainly compared with other face recognition algorithms in a face recognition data set, and is specifically represented as the following table:
the results are shown in the following table:
TABLE 2 YOLO network structure performance test table
In the YOLO v3 of the present embodiment, the input and output sizes of the network layers are different, and the sizes of the network layers are shown in fig. 3. Introduced by the theoretical subsection of the convolutional neural network, the network model is obtained to receive the original pixels of the input video stream image and finally output in the form of probability vectors of categories and coordinates. Inputting an original image into a convolution layer, performing convolution operation, performing a feature matrix based on an activation function sigmoid, and finally extracting image features.
The working principle of YOLO is to segment the input image into S × S meshes, each mesh consisting of (x, y, w, h) and a confidence c (object). Coordinates (x, y) represent the center position of the detection bounding box with respect to the grid. (W, h) is the width and height of the bounding box. If the center of an object falls within a grid cell, the grid cell is responsible for detecting the object. Each cell of the grid predicts a bounding box and a confidence for that box. The calculation formula is as follows:
C(Object)＝Pr(Object)*IOU(Pred，Truth)
where Pr (object) indicates whether the object is contained in the grid. Pr (object) 1 if the mesh contains an object, and pr (object) 0 if the mesh does not contain an object. The IOU (intersection/union) represents the accuracy of the bounding box containing the object, i.e. the overlap ratio of the detected candidate boundary with the ground truth, i.e. the ratio of their intersection/union.
The final confidence level is calculated as follows:
after obtaining the confidence of each prediction box, a low-scoring prediction box is removed by setting a threshold value, and then the rest boundary boxes are subjected to non-maximum inhibition.
The YOLO algorithm uses target detection as a regression problem, using the mean square loss function, and using different weights in different places.
In summary, according to the construction worker identity identification and safety helmet wearing detection method, the construction worker face image is collected, a supervised learning mode is adopted, the Python programming language and the YOLO v3 algorithm are used for extracting the features of the face and the safety helmet, image information is learned and tested, safety helmet wearing detection is achieved, and the Light cnn-29 algorithm is used for identifying the identity of a worker and achieving construction worker identity verification. Through a large amount of data acquisition and training, the accuracy rate of worker identity verification by the method reaches 85%, and the average identification rate of helmet detection reaches more than 90%.
Compared with the traditional face recognition method, the method can recognize the face of a construction worker in various weather and illumination environments, can achieve more than 85% of recognition rate under the condition that people are shielded, and has strong robustness; compared with a general safety helmet detection algorithm, the method can expand the function of worker identity verification on the basis of safety helmet identification, and can realize real-time monitoring.
Correspondingly, the intelligent safety supervision of the construction site is realized. A face information and safety helmet information base of construction workers is established, and a new database is provided for deep learning research in the construction field. Establishing a database, and acquiring front face pictures of safety helmets worn by engineering project department managers and team constructors; and acquiring a monitoring video of a real-name channel entrance, and intercepting worker images under different illumination conditions and various weather environments. For the captured images, we normalized to 416 × 416 size, LabelImg was used to label the worker's facial features and helmet features, and these annotations were saved as XML files in the PASCAL VOC format. In deep learning research, the making of the image label is time-consuming and labor-consuming, and the labeled data set can save much time for later research of scholars.
On the other hand, the embodiment of the invention also discloses a system for identifying the identity of a construction worker and detecting the wearing of a safety helmet, which comprises the following units:
the data collection unit is used for respectively collecting identity information of construction workers and corresponding image information of the safety helmet wearing;
the data processing unit is used for adding labels to the collected image information, storing the labels and performing normalization processing;
the model training unit is used for inputting the image added with the label into a convolutional neural network and training a neural network model;
the model correction unit is used for detecting the performance of the trained convolutional neural network of the trained neural network model, testing the recognition precision and recall rate in various environments and correcting;
and the data detection unit is used for carrying out identity verification on personnel entering the construction site and carrying out safety helmet detection on the personnel entering the construction site by utilizing the corrected neural network model.
It is understood that the system provided by the embodiment of the present invention corresponds to the method provided by the embodiment of the present invention, and the explanation, the example and the beneficial effects of the related contents can refer to the corresponding parts in the method.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A construction worker identity identification and safety helmet wearing detection method is characterized in that:
the method comprises the following steps:
downloading a public data set SHWD, and collecting identity information of construction workers and worker images monitored by a real-name system entrance;
adding a label to the image in the public data set SHWD, naming and storing the collected image by worker identity information, and performing normalization processing;
building a system architecture based on a YOLO V3 and Light cnn-29 neural network algorithm;
training YOLO v3 on the common data set SHWD and inputting the labeled SHWD data set image to YOLO v 3;
taking the image detected as a face area by YOLO v3 as a face recognition training set, and training Light CNN-29;
respectively detecting the performances of a YOLO v3 neural network system and a Light cnn-29 neural network system, testing the recognition accuracy and the recall rate under various construction environments, and correcting;
and verifying the identity of the personnel entering the construction site by using the corrected neural network model, and detecting the safety helmet of the personnel entering the construction site.
2. The construction worker identification and safety helmet wearing detection method according to claim 1, characterized in that: downloading a public data set SHWD and collecting construction worker identity information and worker images monitored by a real-name system entrance, comprising:
first, the department of project manager and each team builder are collected, the face under the condition of wearing the safety helmet is self-photographed, and the photograph information is recorded as follows: putting the name, team and age into a database;
secondly, intercepting facial image information of construction workers through a camera installed at an entrance of a real-name channel of a construction site, expanding a data set, and classifying and storing the data set according to the actual construction site condition;
finally, a common data set SHWD is downloaded, which is the common data set used for headgear wear detection.
3. The construction worker identification and safety helmet wearing detection method according to claim 1, characterized in that: labeling the images in the public data set SHWD, naming and storing the collected images by the identity information of workers, and performing normalization processing, wherein the normalization processing comprises the following steps:
normalizing the collected image to 416 × 416 size, wherein the image content is maximized and comprises human faces and safety helmets;
further comprising:
labeling the classified samples and the data samples labeled with the position boundaries; for the collected data set, LabelImg was used to label the worker's facial features and helmet features, and these annotations were saved as XML files in paschalloc format for Python reading.
4. The construction worker identification and safety helmet wearing detection method according to claim 1, characterized in that: building a system architecture based on a YOLO V3 and Light cnn-29 neural network algorithm, comprising the following steps:
firstly, a system Python environment is built, an image description function is defined, then a trained YOLO v3 and Lightcnn-29 algorithm is introduced, a Haar cascade classifier is introduced, and finally a Utils module is defined, reads the results of the recognition of YOLOv3 and Lightcnn-29, and combines and outputs the results.
5. The construction worker identity identification and safety helmet wearing detection method according to claim 4, characterized in that: the introduced Haar cascade classifier comprises:
the Haar cascade classifier locates the position of the face through an expandable window, and takes the highest score from a prediction window to locate the face.
6. The construction worker identification and safety helmet wearing detection method according to claim 1, characterized in that: training YOLO v3 on the common data set SHWD and inputting labeled SHWD data set images to YOLO v3, comprising:
creating a VOC format training set comprising:
the method comprises the steps of firstly creating an indications folder for storing image marks in an xml format, then creating an ImageSets folder, storing test.txt, train.txt and val.txt in the folders for determining training images and verification images in an SHWD data set, and finally creating a JPEGImages folder for storing all images in the SHWD data set.
7. The construction worker identification and safety helmet wearing detection method according to claim 1, characterized in that: using the image detected as a face region by YOLO v3 as a face recognition training set, training Light CNN-29, including:
the Lightcnn model is implemented using the open source deep learning framework Caffe, Dropout for fully connected layers with a ratio set to 0.7, momentum set to 0.9 and weight attenuation set to 5 × 10 for convolutional layers and fully connected layers except for fc2 layers4；
To avoid overfitting, the fc2 layer was weighted up and attenuated to 5 × 103；
Initial learning rate set to 1 × 103Then gradually decreases to 5 × 105。
8. The construction worker identification and safety helmet wearing detection method according to claim 1, characterized in that: respectively detecting the performances of a YOLO v3 neural network system and a Light cnn-29 neural network system, testing the recognition accuracy and the recall rate under various construction environments, and correcting;
wherein the detecting of the recall rate comprises:
firstly, defining a true case TP, a false positive case FP and a false negative case FN; wherein TP represents the number of construction workers without safety helmets after algorithm operation is identified, FP represents the number of construction workers without safety helmets, but the result is not accurate; FN is the number of misjudged construction workers who do not wear safety helmets; the target identification accuracy represents the proportion of the true sample TP to the total sample TP + FP and is used for measuring the reliability of identification performance;
the recall rate represents the proportion of true positive samples TP to total positive samples (TP + FN);
the specific calculation formula is as follows:
9. the construction worker identification and safety helmet wearing detection method according to claim 8, characterized in that: detecting the performance of the trained convolutional neural network of the trained neural network model, testing the recognition precision and recall rate in various environments, and correcting;
wherein the detection of the accuracy of the test identification under various environments comprises:
different weather environment and same day's different time, illumination all can change, can all produce the influence to the definition of image, and the change of number of people all can produce the influence to the recognition accuracy of algorithm, takes these two items into account, and concrete classification is shown as the following table:
the expression of face recognition is mainly compared with other face recognition algorithms in a face recognition data set, and is specifically represented as the following table:
10. the utility model provides a detection system is worn to construction worker identification and safety helmet which characterized in that:
the method comprises the following units:
the data collection unit is used for respectively collecting identity information of construction workers and corresponding image information of the safety helmet wearing;
the data processing unit is used for adding labels to the collected image information, storing the labels and performing normalization processing;
the model training unit is used for inputting the image added with the label into a convolutional neural network and training a neural network model;
the model correction unit is used for detecting the performance of the trained convolutional neural network of the trained neural network model, testing the recognition precision and recall rate in various environments and correcting;
and the data detection unit is used for carrying out identity verification on personnel entering the construction site and carrying out safety helmet detection on the personnel entering the construction site by utilizing the corrected neural network model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010447288.2A CN111598040B (en) | 2020-05-25 | 2020-05-25 | Construction worker identity recognition and safety helmet wearing detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010447288.2A CN111598040B (en) | 2020-05-25 | 2020-05-25 | Construction worker identity recognition and safety helmet wearing detection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111598040A true CN111598040A (en) | 2020-08-28 |
| CN111598040B CN111598040B (en) | 2024-05-14 |
Family
ID=72191399
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010447288.2A Active CN111598040B (en) | 2020-05-25 | 2020-05-25 | Construction worker identity recognition and safety helmet wearing detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111598040B (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112036360A (en) * | 2020-09-10 | 2020-12-04 | 杭州云栖智慧视通科技有限公司 | Method for identifying attributes of helmet of rider |
| CN112183284A (en) * | 2020-09-22 | 2021-01-05 | 上海钧正网络科技有限公司 | Safety information verification and designated driving order receiving control method and device |
| CN112347943A (en) * | 2020-11-09 | 2021-02-09 | 哈尔滨理工大学 | Anchor optimization safety helmet detection method based on YOLOV4 |
| CN112396658A (en) * | 2020-11-30 | 2021-02-23 | 同济人工智能研究院(苏州)有限公司 | Indoor personnel positioning method and positioning system based on video |
| CN112434828A (en) * | 2020-11-23 | 2021-03-02 | 南京富岛软件有限公司 | Intelligent identification method for safety protection in 5T operation and maintenance |
| CN112672052A (en) * | 2020-12-24 | 2021-04-16 | 华中光电技术研究所（中国船舶重工集团公司第七一七研究所） | Image data enhancement method and system, electronic equipment and storage medium |
| CN112668465A (en) * | 2020-12-25 | 2021-04-16 | 秒影工场(北京)科技有限公司 | Film face extraction method based on multistage CNN |
| CN112818175A (en) * | 2021-02-07 | 2021-05-18 | 中国矿业大学 | Factory worker searching method and training method of worker recognition model |
| CN113256934A (en) * | 2021-05-18 | 2021-08-13 | 哈尔滨理工大学 | Safety detection system for aerial work personnel |
| CN113516076A (en) * | 2021-07-12 | 2021-10-19 | 大连民族大学 | Improved lightweight YOLO v4 safety protection detection method based on attention mechanism |
| CN114283485A (en) * | 2022-03-04 | 2022-04-05 | 杭州格物智安科技有限公司 | Safety helmet wearing detection method and device, storage medium and safety helmet |
| CN114511427A (en) * | 2022-04-21 | 2022-05-17 | 四川省大数据中心 | Safety education supervision method, device and system for project site |
| CN114627499A (en) * | 2022-03-07 | 2022-06-14 | 上海应用技术大学 | Online safety helmet face recognition method based on convolutional neural network |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190050632A1 (en) * | 2017-08-14 | 2019-02-14 | Baidu Online Network Technology (Beijing) Co., Ltd . | Method and apparatus for generating training data for human face recognition, device and computer storage medium |
| CN110119686A (en) * | 2019-04-17 | 2019-08-13 | 电子科技大学 | A kind of safety cap real-time detection method based on convolutional neural networks |
| CN110309719A (en) * | 2019-05-27 | 2019-10-08 | 安徽继远软件有限公司 | A kind of electric network operation personnel safety cap wears management control method and system |
| CN110399905A (en) * | 2019-07-03 | 2019-11-01 | 常州大学 | The detection and description method of safety cap wear condition in scene of constructing |
| CN110533811A (en) * | 2019-08-28 | 2019-12-03 | 深圳市万睿智能科技有限公司 | The method and device and system and storage medium of safety cap inspection are realized based on SSD |
| CN110728223A (en) * | 2019-10-08 | 2020-01-24 | 济南东朔微电子有限公司 | Helmet wearing identification method based on deep learning |
| CN110738127A (en) * | 2019-09-19 | 2020-01-31 | 福建师范大学福清分校 | Helmet identification method based on unsupervised deep learning neural network algorithm |
| CN111160440A (en) * | 2019-12-24 | 2020-05-15 | 广东省智能制造研究所 | Helmet wearing detection method and device based on deep learning |
-
2020
- 2020-05-25 CN CN202010447288.2A patent/CN111598040B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190050632A1 (en) * | 2017-08-14 | 2019-02-14 | Baidu Online Network Technology (Beijing) Co., Ltd . | Method and apparatus for generating training data for human face recognition, device and computer storage medium |
| CN110119686A (en) * | 2019-04-17 | 2019-08-13 | 电子科技大学 | A kind of safety cap real-time detection method based on convolutional neural networks |
| CN110309719A (en) * | 2019-05-27 | 2019-10-08 | 安徽继远软件有限公司 | A kind of electric network operation personnel safety cap wears management control method and system |
| CN110399905A (en) * | 2019-07-03 | 2019-11-01 | 常州大学 | The detection and description method of safety cap wear condition in scene of constructing |
| CN110533811A (en) * | 2019-08-28 | 2019-12-03 | 深圳市万睿智能科技有限公司 | The method and device and system and storage medium of safety cap inspection are realized based on SSD |
| CN110738127A (en) * | 2019-09-19 | 2020-01-31 | 福建师范大学福清分校 | Helmet identification method based on unsupervised deep learning neural network algorithm |
| CN110728223A (en) * | 2019-10-08 | 2020-01-24 | 济南东朔微电子有限公司 | Helmet wearing identification method based on deep learning |
| CN111160440A (en) * | 2019-12-24 | 2020-05-15 | 广东省智能制造研究所 | Helmet wearing detection method and device based on deep learning |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112036360A (en) * | 2020-09-10 | 2020-12-04 | 杭州云栖智慧视通科技有限公司 | Method for identifying attributes of helmet of rider |
| CN112036360B (en) * | 2020-09-10 | 2023-11-28 | 杭州云栖智慧视通科技有限公司 | Riding helmet attribute identification method |
| CN112183284B (en) * | 2020-09-22 | 2022-09-23 | 上海钧正网络科技有限公司 | Safety information verification and designated driving order receiving control method and device |
| CN112183284A (en) * | 2020-09-22 | 2021-01-05 | 上海钧正网络科技有限公司 | Safety information verification and designated driving order receiving control method and device |
| CN112347943A (en) * | 2020-11-09 | 2021-02-09 | 哈尔滨理工大学 | Anchor optimization safety helmet detection method based on YOLOV4 |
| CN112434828A (en) * | 2020-11-23 | 2021-03-02 | 南京富岛软件有限公司 | Intelligent identification method for safety protection in 5T operation and maintenance |
| CN112434828B (en) * | 2020-11-23 | 2023-05-16 | 南京富岛软件有限公司 | Intelligent safety protection identification method in 5T operation and maintenance |
| CN112396658A (en) * | 2020-11-30 | 2021-02-23 | 同济人工智能研究院(苏州)有限公司 | Indoor personnel positioning method and positioning system based on video |
| CN112396658B (en) * | 2020-11-30 | 2024-03-19 | 同济人工智能研究院(苏州)有限公司 | Indoor personnel positioning method and system based on video |
| CN112672052A (en) * | 2020-12-24 | 2021-04-16 | 华中光电技术研究所（中国船舶重工集团公司第七一七研究所） | Image data enhancement method and system, electronic equipment and storage medium |
| CN112668465A (en) * | 2020-12-25 | 2021-04-16 | 秒影工场(北京)科技有限公司 | Film face extraction method based on multistage CNN |
| CN112818175B (en) * | 2021-02-07 | 2023-09-01 | 中国矿业大学 | Factory staff searching method and training method of staff identification model |
| CN112818175A (en) * | 2021-02-07 | 2021-05-18 | 中国矿业大学 | Factory worker searching method and training method of worker recognition model |
| CN113256934A (en) * | 2021-05-18 | 2021-08-13 | 哈尔滨理工大学 | Safety detection system for aerial work personnel |
| CN113516076A (en) * | 2021-07-12 | 2021-10-19 | 大连民族大学 | Improved lightweight YOLO v4 safety protection detection method based on attention mechanism |
| CN113516076B (en) * | 2021-07-12 | 2023-09-01 | 大连民族大学 | Attention mechanism improvement-based lightweight YOLO v4 safety protection detection method |
| CN114283485A (en) * | 2022-03-04 | 2022-04-05 | 杭州格物智安科技有限公司 | Safety helmet wearing detection method and device, storage medium and safety helmet |
| CN114283485B (en) * | 2022-03-04 | 2022-10-14 | 杭州格物智安科技有限公司 | Safety helmet wearing detection method and device, storage medium and safety helmet |
| CN114627499A (en) * | 2022-03-07 | 2022-06-14 | 上海应用技术大学 | Online safety helmet face recognition method based on convolutional neural network |
| CN114627499B (en) * | 2022-03-07 | 2024-04-09 | 上海应用技术大学 | On-line safety helmet face recognition method based on convolutional neural network |
| CN114511427A (en) * | 2022-04-21 | 2022-05-17 | 四川省大数据中心 | Safety education supervision method, device and system for project site |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111598040B (en) | 2024-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111598040B (en) | Construction worker identity recognition and safety helmet wearing detection method and system | |
| CN112200043B (en) | Intelligent danger source identification system and method for outdoor construction site | |
| CN108319926A (en) | A kind of the safety cap wearing detecting system and detection method of building-site | |
| CN113516076B (en) | Attention mechanism improvement-based lightweight YOLO v4 safety protection detection method | |
| CN110807429A (en) | Construction safety detection method and system based on tiny-YOLOv3 | |
| CN110674702B (en) | Mine image scene classification method, device, equipment and system | |
| CN112613454A (en) | Electric power infrastructure construction site violation identification method and system | |
| CN112347943A (en) | Anchor optimization safety helmet detection method based on YOLOV4 | |
| CN111652046A (en) | Safe wearing detection method, equipment and system based on deep learning | |
| CN117035419B (en) | Intelligent management system and method for enterprise project implementation | |
| CN112184773A (en) | Helmet wearing detection method and system based on deep learning | |
| CN113807240A (en) | Intelligent transformer substation personnel dressing monitoring method based on uncooperative face recognition | |
| CN113191273A (en) | Oil field well site video target detection and identification method and system based on neural network | |
| CN115512387A (en) | Construction site safety helmet wearing detection method based on improved YOLOV5 model | |
| CN117726991B (en) | High-altitude hanging basket safety belt detection method and terminal | |
| CN117576632B (en) | Multi-mode AI large model-based power grid monitoring fire early warning system and method | |
| CN113052125B (en) | Construction site violation image recognition and alarm method | |
| CN116311081B (en) | Medical laboratory monitoring image analysis method and system based on image recognition | |
| CN115311601A (en) | Fire detection analysis method based on video analysis technology | |
| Wu et al. | Design and implementation of safety helmet detection system based on computer vision | |
| Aleksyuk et al. | Application of the Yolov 8 Neural Network for Detection of Personal Protection Equipment in Hazardous Enterprises | |
| CN117475471A (en) | Protective article wearing detection method based on improved YOLOv7 | |
| Tang et al. | Real-time Detection of Safety Helmet and Workwear Based on YOLOv7 | |
| CN118135490A (en) | Potential safety hazard investigation and accident early warning method and system based on graphic configuration | |
| CN118195582A (en) | Operation and maintenance abnormal state prediction method applied to pollution source operation and maintenance platform |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Sun Xinya Inventor after: Xiong Haohan Inventor after: Hu Jindong Inventor after: Ge Ling Inventor after: Jiang Xing Inventor after: Jiang Xiaoyan Inventor after: Chu Ruibing Inventor after: Chen Shuanggui Inventor after: Yu Xi Inventor after: Zhou Wenwu Inventor after: Shu Bin Inventor after: Song Xiaoli Inventor after: Chen Liang Inventor before: Sun Xinya Inventor before: Hu Jindong Inventor before: Ge Ling Inventor before: Jiang Xing Inventor before: Chu Ruibing Inventor before: Chen Shuanggui Inventor before: Yu Xi Inventor before: Zhou Wenwu Inventor before: Shu Bin Inventor before: Song Xiaoli Inventor before: Chen Liang Inventor before: Xiong Haohan |