CN111652046A - Safe wearing detection method, equipment and system based on deep learning - Google Patents
Safe wearing detection method, equipment and system based on deep learning Download PDFInfo
- Publication number
- CN111652046A CN111652046A CN202010306273.4A CN202010306273A CN111652046A CN 111652046 A CN111652046 A CN 111652046A CN 202010306273 A CN202010306273 A CN 202010306273A CN 111652046 A CN111652046 A CN 111652046A
- Authority
- CN
- China
- Prior art keywords
- image
- safety
- deep learning
- wearing
- personnel
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 47
- 238000013135 deep learning Methods 0.000 title claims abstract description 18
- 238000000034 method Methods 0.000 claims abstract description 48
- 238000013136 deep learning model Methods 0.000 claims abstract description 40
- 238000004519 manufacturing process Methods 0.000 claims abstract description 24
- 238000000605 extraction Methods 0.000 claims abstract description 15
- 238000012549 training Methods 0.000 claims description 45
- 238000013528 artificial neural network Methods 0.000 claims description 17
- 238000012360 testing method Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 11
- 238000012795 verification Methods 0.000 claims description 11
- 238000005457 optimization Methods 0.000 claims description 8
- 238000012544 monitoring process Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 6
- 239000000126 substance Substances 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 abstract description 3
- 238000011897 real-time detection Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 13
- 230000008569 process Effects 0.000 description 11
- 238000003860 storage Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 238000013473 artificial intelligence Methods 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 241001292396 Cirrhitidae Species 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000001681 protective effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000036512 infertility Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000003068 static effect Effects 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
- 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
- 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
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
- G06V40/103—Static body considered as a whole, e.g. static pedestrian or occupant recognition
-
- 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
- G06V40/161—Detection; Localisation; Normalisation
-
- 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
Abstract
The embodiment of the invention relates to a safety wearing detection method, equipment and a system based on deep learning, wherein the method comprises the following steps: receiving an image of a production operation field sent by image acquisition equipment; performing feature extraction on the image based on a deep learning model to obtain a plurality of safety wearing features of the personnel in the image; and determining the safety wearing condition of the personnel by judging whether the plurality of safety wearing characteristics accord with preset values. According to the embodiment of the invention, the recognition accuracy of the safety wearing characteristics of the working personnel is improved through the convolutional neural network deep learning algorithm, and the detection target (the safety wearing characteristics) can be expanded. Carry out real-time detection to the staff at the production operation in-process, the wearing condition of discernment staff safety equipment to improve the operation personnel and dress detection efficiency, strengthen staff's safety in production management and control.
Description
Technical Field
The invention relates to the technical field of computer vision and artificial intelligence, in particular to a method, equipment and a system for processing scenic spot tourist photos based on face recognition.
Background
In the production operation process of each industry, the safety of the staff is paid much attention. In order to ensure the safety of the staff in the production construction process, the staff can wear various standard protective equipment, and the staff needs to be supervised by a supervisor in the whole process. However, some production workshops have high requirements on the environment, for example, sterility, dust-free and the like are needed, and supervisors need to wear corresponding protective equipment to check the safety wearing conditions of workers to enter the workshops, so that the time cost is high.
Disclosure of Invention
The embodiment of the invention aims to solve the following technical problems at least to a certain extent:
confirm that staff's safe wearing situation passes through the manual work, and is with high costs.
The first aspect of the embodiment of the invention provides a safety wearing detection method based on deep learning, which comprises the following steps:
receiving an image of a production operation field sent by image acquisition equipment;
performing feature extraction on the image based on a deep learning model to obtain a plurality of safety wearing features of the personnel in the image;
and determining the safety wearing condition of the personnel by judging whether the plurality of safety wearing characteristics accord with preset values.
In one example, the plurality of safety wear features includes at least one or more of a hard hat, a mask, a bracelet, gloves, a workwear, and a workcard.
In one example, the determining the safety wearing condition of the person by determining whether the plurality of safety wearing characteristics meet preset values includes:
if the plurality of safety wearing characteristics do not accord with preset values, acquiring the face characteristic information of the personnel in the image through a deep learning model;
and matching the face characteristic information with face data in a database to determine the identity of the person in the image.
In one example, further comprising:
if the plurality of safety wearing characteristics do not accord with preset values, alarming is carried out in the system, and
and sending the identity of the personnel in the image and the safety wearing condition of the personnel to a monitoring party.
In one example, further comprising:
receiving feedback of the supervisor;
and according to the feedback of the supervisor, putting the image into a data set of the deep learning model, and performing optimization training on the deep learning model.
In one example, the performing feature extraction on the image based on the deep learning model to obtain a plurality of safety wearing features of the person in the image includes:
setting the deep learning model to identify the type of the safety wearing feature;
and acquiring a plurality of safety wearing characteristics of the personnel in the image according to the type of the safety wearing characteristics.
In one example, the performing feature extraction on the image based on the deep learning model to obtain a plurality of safety wearing features of the person in the image includes:
preprocessing the image through specified parameters, wherein the specified parameters comprise the size of the image and the number, type and numerical range of channels;
and performing feature extraction on the image through the deep learning model to obtain a plurality of safety wearing features of the personnel in the image.
In one example, before the receiving the image of the production job site sent by the image acquisition device, the method further includes: training the deep learning model to a target accuracy, wherein the training step comprises:
acquiring the marked image data, and summarizing the acquired image data into an original data set;
processing the original data set by an operator of the neural network through preset parameters, and dividing the preprocessed original data set into a training set, a verification set and a test set, wherein the preset parameters comprise: the size of the image; number of channels, type, and value range;
setting hyper-parameters of a neural network, substituting the training set, the verification set and the test set into the neural network, training and parameter solving the neural network by a nonlinear optimization algorithm, and performing iterative training until the neural network reaches a target accuracy.
A second aspect of an embodiment of the present invention provides a safety wearing detection device based on deep learning, including: at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to:
receiving an image of a production operation field sent by image acquisition equipment;
performing feature extraction on the image based on a deep learning model to obtain a plurality of safety wearing features of the personnel in the image;
and determining the safety wearing condition of the personnel by judging whether the plurality of safety wearing characteristics accord with preset values.
A third aspect of an embodiment of the present invention provides a safety wearing detection system based on deep learning, including:
the image acquisition equipment is used for acquiring an image of a production operation site;
the safety wearing detection equipment is connected with the image acquisition equipment and is used for receiving the image sent by the image acquisition equipment and determining the safety wearing condition of personnel in the image based on a deep learning model;
and the display equipment is connected with the safety wearing detection equipment and is used for displaying the safety wearing condition of the personnel in the image.
Has the advantages that:
according to the embodiment of the invention, the recognition accuracy of the safety wearing characteristics of the working personnel is improved through the convolutional neural network deep learning algorithm, and the detection target (the safety wearing characteristics) can be expanded. Setting an equipment list to be detected in the system, automatically checking the system before a worker enters a construction area, and correctly identifying the equipment after the worker correctly wears the specified equipment according to the detection list; carry out real-time detection to the staff at the production operation in-process, the wearing condition of discernment staff safety equipment to improve the operation personnel and dress detection efficiency, strengthen staff's safety in production management and control.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
FIG. 1 is a schematic flow chart of a method provided by an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a training method of an artificial intelligence model according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an apparatus framework provided by an embodiment of the present invention;
fig. 4 is a schematic diagram of a system framework according to an embodiment of the present invention.
Detailed Description
In order to more clearly explain the overall concept of the present application, the following detailed description is given by way of example in conjunction with the accompanying drawings.
The embodiment of the invention provides a safety wearing detection method based on deep learning and a corresponding scheme, which improve the identification accuracy of safety wearing characteristics of workers through a convolutional neural network deep learning algorithm and can expand a detection target (safety wearing characteristics). Setting an equipment list to be detected in the system, automatically checking the system before a worker enters a construction area, and correctly identifying the equipment after the worker correctly wears the specified equipment according to the detection list; carry out real-time detection to the staff at the production operation in-process, the wearing condition of discernment staff safety equipment to improve the operation personnel and dress detection efficiency, strengthen staff's safety in production management and control.
Some embodiments of the invention are described in detail below with reference to the accompanying drawings.
According to a first aspect of the embodiments of the present invention, a method for processing a photo of a tourist in a scenic region based on face recognition is provided, fig. 1 is a schematic flow chart of the method of the first aspect of the embodiments of the present invention, and the method is applicable to an image processing device or an image processing server, and as shown in the figure, the method includes:
s101, receiving an image of a production operation field sent by image acquisition equipment;
s102, extracting the features of the image based on a deep learning model, and acquiring a plurality of safety wearing features of the personnel in the image;
s103, determining the safety wearing condition of the personnel by judging whether the plurality of safety wearing characteristics accord with preset values.
According to the specific embodiment of the invention, the image acquisition equipment comprises a plurality of network cameras, each network camera can acquire images from multiple angles by setting a scanning range, and the network cameras can cover all production operation sites. In some preferred embodiments of the present invention, the image capturing device extracts a key frame image from the acquired real-time image, and sends the key frame image to the safety wearing detection device.
According to the embodiment of the present invention, the safety wear characteristics of the safety personnel may be one or more of a safety helmet, a mask, a bracelet, gloves, an operation suit and a work card in a common working environment, and may also be radiation protection suits, goggles and the like in an uncommon working environment, which is not particularly limited by the embodiment of the present invention.
According to the specific embodiment of the present invention, before performing feature extraction on the image based on the deep learning model, the method further includes preprocessing the image by using specified parameters in step S102, where the specified parameters include image size, number, type, and value range of channels; and performing feature extraction on the image through the deep learning model to obtain a plurality of safety wearing features of the personnel in the image. For a specific implementation, the following description may refer to an image preprocessing process during deep learning model training, which is not described herein again.
According to an embodiment of the present invention, in step S102, the performing feature extraction on the image based on the deep learning model to obtain a plurality of safety wearing features of the person in the image may further include: and setting the deep learning model to identify the type of the safety wearing feature, for example, in an unusual working environment, setting the type of the safety wearing feature to be radiation-proof clothes and goggles, and acquiring a plurality of safety wearing features of personnel in the image according to the type of the safety wearing feature.
According to an embodiment of the present invention, in step S103, the determining whether the plurality of safety wearing features meet the preset values may be interpreted in a broad sense, for example, by setting the preset values to 0 and 1, i.e., determining whether all of the plurality of safety wearing features are included.
In some preferred embodiments of the present invention, the determining the safety wearing condition of the person by determining whether the plurality of safety wearing characteristics meet preset values includes: if the plurality of safety wearing characteristics do not accord with preset values, acquiring the face characteristic information of the personnel in the image through a deep learning model; and matching the face characteristic information with face data in a database to determine the identity of the person in the image.
It can be understood that the artificial intelligence model for face recognition may be trained by the artificial intelligence model training method disclosed in the embodiment of the present invention, or may be a human face recognition model trained by a third-party platform.
In other preferred embodiments of the present invention, further comprising: and if the plurality of safety wearing characteristics do not accord with preset values, giving an alarm in the system, and sending the identity of the personnel in the image and the safety wearing condition of the personnel to a monitoring party.
Correspondingly, the detection equipment can receive the feedback of the monitoring party, put the image into the data set of the deep learning model and carry out optimization training on the deep learning model. For example, if the monitoring party feeds back that the alarm is correct, the picture is put into a verification set of the data set, and if the monitoring party feeds back that the alarm is not completely correct or wrong, the picture is put into a training set of the data set.
Fig. 2 is a schematic diagram of a training process of an artificial intelligence model according to an embodiment of the present invention, and as shown in fig. 2, the training process includes:
s201, acquiring the image data after the labeling, and summarizing the acquired image data into an original data set;
s202, the operator of the neural network processes the original data set through preset parameters, and divides the preprocessed original data set into a training set, a verification set and a test set, wherein the preset parameters comprise: the size of the image; number of channels, type, and value range;
s203, setting hyper-parameters of the neural network, substituting the training set, the verification set and the test set into the neural network, training and parameter solving the neural network by a nonlinear optimization algorithm, and performing iterative training until the neural network reaches the target accuracy.
In some embodiments of the present invention, the deep learning model is trained and implemented by a Halcon deep learning algorithm, and the specific steps are as follows.
Firstly, reading and reading image data input by a user, and classifying and marking the detected characteristics on the image data by the user through a classifying and marking function.
And reading the data set in the specified path by using a read _ dl _ classifier _ data _ set method, and acquiring the original data set with the label. Reading a pre-training network in a mode of reading _ DL _ classifier, reading 'pre-trained _ DL _ classifier _ compact.hdl' and 'pre-trained _ DL _ classifier _ enhanced.hdl' provided by a Halcon deep learning algorithm, preprocessing image data in an original data set, preprocessing an input image into an image with consistent parameters including image size, channel number, type and value range, and removing an image background and a part irrelevant to the characteristics. Specifically, the operator threshold is used to perform threshold segmentation on the image to obtain a connected domain of the entity, the minimum circumscribed rectangle of the connected domain is obtained, and then the operator reduce _ domain is used to segment the detection object from the source image. And finally, converting the segmented image into a real type by using an operator convert _ image _ type, and scaling the image into 255 x 255 pixels by using a scale _ image operator to obtain input data required by the classifier.
And (3) dividing the read preprocessed data set into a training set, a verification set and a test set by using a split _ dl _ classifier _ data _ set method. The proportion of the training set is 70%, the training set is used for model training, the verification set is used for improving the accuracy of model training, and the test set is used for evaluating the trained model.
And setting a model training super parameter, and retraining the neural network by setting a label, a batch processing size, a running environment, an initial learning rate, a periodicity, a regularization parameter and other super parameters through an operator set _ dl _ classifier _ param.
And (3) using a training set, performing sample learning by using an operator train _ dl _ classifier _ batch, and performing model training and parameter solving by using a nonlinear optimization algorithm to minimize the value of a loss function to obtain an optimal solution. The nonlinear optimization algorithm uses a momentum-based random gradient descent (SGD) algorithm, namely, network errors are calculated and propagated backwards during each batch learning, and network parameters are updated according to gradient information. Increasing the number of learning cycles and the number of iterations, reducing the error rate of training and verification, evaluating the performance of classifiers in a neural network in a current training set and a verification set by using application _ dl _ classifier and evaluation _ dl _ classifier methods, storing the trained classifiers by using write _ dl _ classifier methods, and completing model training.
And finally, applying the classifier to the test set, using the test set to verify the training model, firstly reading the trained classifier by using a read _ dl _ classifier method, then calling an application _ dl _ classifier _ base method to apply the trained classifier to the test set, and then calling an evaluation _ dl _ classifier method to evaluate the classification result in the test set. And predicting the matching degree of the category and the real label, and meeting the requirement on the accuracy of the training data set.
Based on the same idea, some embodiments of the present application further provide a device and a system corresponding to the above method.
Fig. 3 is a schematic structural diagram of a frame of the deep learning-based safety wearable detection device shown in fig. 1 according to an embodiment of the present invention, and as shown in fig. 3, the safety wearable detection device includes: at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to:
receiving an image of a production operation field sent by image acquisition equipment;
performing feature extraction on the image based on a deep learning model to obtain a plurality of safety wearing features of the personnel in the image;
and determining the safety wearing condition of the personnel by judging whether the plurality of safety wearing characteristics accord with preset values.
Fig. 4 is a schematic structural diagram of a deep learning based safety wearing detection system corresponding to fig. 1, as shown in fig. 4, the system includes:
the image acquisition equipment is used for acquiring an image of a production operation site;
the safety wearing detection equipment is connected with the image acquisition equipment and is used for receiving the image sent by the image acquisition equipment and determining the safety wearing condition of personnel in the image based on a deep learning model;
and the display equipment is connected with the safety wearing detection equipment and used for displaying the safety wearing condition of the personnel in the image and sending feedback to the safety wearing equipment.
The image acquisition equipment comprises a plurality of network cameras, each network camera can acquire images from multiple angles by setting a scanning range, and the network cameras can cover all production operation sites. In some preferred embodiments of the present invention, the image capturing device extracts a key frame image from the acquired real-time image, and sends the key frame image to the safety wearing detection device.
The safety wearing detection equipment can be a background server and is connected with a plurality of image acquisition terminals and comprises an image acquisition module, a deep learning model training module, a deep learning model characteristic detection module, a database and a display module.
The image acquisition module is connected to the image acquisition terminal through a network, receives the key frame image, preprocesses the image and is used for characteristic analysis.
The deep learning model training module is used for setting target detection characteristics, including face detection and recognition and object detection. For face detection and recognition, the module has a face detection model, and can input a face picture and a corresponding person name. The deep learning model training module is preset with a plurality of feature detection models, supports setting and adding a plurality of detection features, such as a safety helmet, a mask, a bracelet, gloves, an operating suit, a workcard and the like, provides a plurality of detection sample images for each detection feature and performs detection mark calibration operation. And the detection model training module performs feature training on each feature by using a deep learning algorithm to generate a feature point recognition model. And updating the recognition model according to the subsequently added detection samples, and accelerating the characteristic detection speed and accuracy.
The deep learning model feature detection module detects each picture based on a model detection algorithm, and when a certain detection target is not identified, the corresponding person name is acquired through face identification, and an alarm is generated and displayed in the system display module.
The display module provides system configuration, model training entry and discernment show function, can show video monitoring area all staff's safety equipment wearing condition, and the notice information includes personnel's discernment name, safe real-time status of wearing, provides safety supervision personnel.
The embodiments in the present application are described in a progressive manner, and the same and similar parts among the embodiments can be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for the device and media embodiments, since they are substantially similar to the platform embodiments, the description is relatively simple, and reference may be made to some descriptions of the platform embodiments for relevant points.
The device and the system provided by the embodiment of the application correspond to the method, so the device and the system also have the similar beneficial technical effects as the corresponding method, and the beneficial technical effects of the platform are explained in detail above, so the beneficial technical effects of the device and the medium are not repeated herein.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention 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 invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. 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.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A safety wearing detection method based on deep learning is characterized by comprising the following steps:
receiving an image of a production operation field sent by image acquisition equipment;
performing feature extraction on the image based on a deep learning model to obtain a plurality of safety wearing features of the personnel in the image;
and determining the safety wearing condition of the personnel by judging whether the plurality of safety wearing characteristics accord with preset values.
2. The method of claim 1, wherein the plurality of safety wear features includes at least one or more of a safety helmet, a mask, a bracelet, a glove, a workwear, and a workcard.
3. The method of claim 1, wherein determining the safety wear condition of the person by determining whether the plurality of safety wear features meet a predetermined value comprises:
if the plurality of safety wearing characteristics do not accord with preset values, acquiring the face characteristic information of the personnel in the image through a deep learning model;
and matching the face characteristic information with face data in a database to determine the identity of the person in the image.
4. The method of claim 3, further comprising:
if the plurality of safety wearing characteristics do not accord with preset values, alarming is carried out in the system, and
and sending the identity of the personnel in the image and the safety wearing condition of the personnel to a monitoring party.
5. The method of claim 4, further comprising:
receiving feedback of the supervisor;
and according to the feedback of the supervisor, putting the image into a data set of the deep learning model, and performing optimization training on the deep learning model.
6. The method of claim 1, wherein the performing feature extraction on the image based on the deep learning model to obtain a plurality of safety wearing features of the person in the image comprises:
setting the deep learning model to identify the type of the safety wearing feature;
and acquiring a plurality of safety wearing characteristics of the personnel in the image according to the type of the safety wearing characteristics.
7. The method of claim 1, wherein the performing feature extraction on the image based on the deep learning model to obtain a plurality of safety wearing features of the person in the image comprises:
preprocessing the image through specified parameters, wherein the specified parameters comprise the size of the image and the number, type and numerical range of channels;
and performing feature extraction on the image through the deep learning model to obtain a plurality of safety wearing features of the personnel in the image.
8. The method of claim 1, wherein the receiving the image of the production job site sent by the image capture device further comprises: training the deep learning model to a target accuracy, wherein the training step comprises:
acquiring the marked image data, and summarizing the acquired image data into an original data set;
processing the original data set by an operator of the neural network through preset parameters, and dividing the preprocessed original data set into a training set, a verification set and a test set, wherein the preset parameters comprise: the size of the image; number of channels, type, and value range;
setting hyper-parameters of a neural network, substituting the training set, the verification set and the test set into the neural network, training and parameter solving the neural network by a nonlinear optimization algorithm, and performing iterative training until the neural network reaches a target accuracy.
9. The utility model provides a safe check out test set of wearing based on deep learning which characterized in that includes: at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to:
receiving an image of a production operation field sent by image acquisition equipment;
performing feature extraction on the image based on a deep learning model to obtain a plurality of safety wearing features of the personnel in the image;
and determining the safety wearing condition of the personnel by judging whether the plurality of safety wearing characteristics accord with preset values.
10. The utility model provides a safe detecting system of wearing based on deep learning which characterized in that includes:
the image acquisition equipment is used for acquiring an image of a production operation site;
the safety wearing detection equipment is connected with the image acquisition equipment and is used for receiving the image sent by the image acquisition equipment and determining the safety wearing condition of personnel in the image based on a deep learning model;
and the display equipment is connected with the safety wearing detection equipment and is used for displaying the safety wearing condition of the personnel in the image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010306273.4A CN111652046A (en) | 2020-04-17 | 2020-04-17 | Safe wearing detection method, equipment and system based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010306273.4A CN111652046A (en) | 2020-04-17 | 2020-04-17 | Safe wearing detection method, equipment and system based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111652046A true CN111652046A (en) | 2020-09-11 |
Family
ID=72342970
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010306273.4A Pending CN111652046A (en) | 2020-04-17 | 2020-04-17 | Safe wearing detection method, equipment and system based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111652046A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112183264A (en) * | 2020-09-17 | 2021-01-05 | 国网天津静海供电有限公司 | Method for judging people lingering under crane boom based on spatial relationship learning |
| CN112507945A (en) * | 2020-12-18 | 2021-03-16 | 南方电网电力科技股份有限公司 | Method and device for managing and controlling behavior of operator, electronic equipment and storage medium |
| CN114882442A (en) * | 2022-05-31 | 2022-08-09 | 广州信诚信息科技有限公司 | Personnel and equipment situation identification method based on electric power operation site |
| CN115601712A (en) * | 2022-12-15 | 2023-01-13 | 南京电力自动化设备三厂有限公司（Cn） | Image data processing method and system suitable for field safety measures |
| CN116958702A (en) * | 2023-08-01 | 2023-10-27 | 浙江钛比科技有限公司 | Hotel guard personnel wearing detection method and system based on edge artificial intelligence |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106295601A (en) * | 2016-08-18 | 2017-01-04 | 合肥工业大学 | A kind of Safe belt detection method of improvement |
| CN107680220A (en) * | 2017-09-28 | 2018-02-09 | 朱明增 | A kind of operation unlawful practice intelligent identification Method based on machine vision technique |
| CN107909034A (en) * | 2017-11-15 | 2018-04-13 | 清华大学深圳研究生院 | A kind of method for detecting human face, device and computer-readable recording medium |
| US20180204111A1 (en) * | 2013-02-28 | 2018-07-19 | Z Advanced Computing, Inc. | System and Method for Extremely Efficient Image and Pattern Recognition and Artificial Intelligence Platform |
| US20180333083A1 (en) * | 2015-02-25 | 2018-11-22 | Leonardo Y. Orellano | Fall detection systems and methods |
| CN109543542A (en) * | 2018-10-24 | 2019-03-29 | 杭州叙简科技股份有限公司 | A kind of determination method whether particular place personnel dressing standardizes |
| CN109635697A (en) * | 2018-12-04 | 2019-04-16 | 国网浙江省电力有限公司电力科学研究院 | Electric operating personnel safety dressing detection method based on YOLOv3 target detection |
| CN109635758A (en) * | 2018-12-18 | 2019-04-16 | 武汉市蓝领英才科技有限公司 | Wisdom building site detection method is dressed based on the high altitude operation personnel safety band of video |
| CN109757791A (en) * | 2019-01-25 | 2019-05-17 | 武汉职业技术学院 | A kind of intelligent and safe school uniform device |
| CN110070055A (en) * | 2019-04-25 | 2019-07-30 | 南京邮电大学 | A kind of capital construction scene safety detecting system and method based on deep learning |
| CN110119656A (en) * | 2018-02-07 | 2019-08-13 | 中国石油化工股份有限公司 | Intelligent monitor system and the scene monitoring method violating the regulations of operation field personnel violating the regulations |
| CN110309768A (en) * | 2019-06-28 | 2019-10-08 | 上海眼控科技股份有限公司 | The staff's detection method and equipment of car test station |
| CN110334619A (en) * | 2019-06-21 | 2019-10-15 | 上海翊视皓瞳信息科技有限公司 | A kind of secure supervision system and method based on wearable device |
| 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 |
| CN110674772A (en) * | 2019-09-29 | 2020-01-10 | 国家电网有限公司技术学院分公司 | Intelligent safety control auxiliary system and method for electric power operation site |
| 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 |
| CN110795989A (en) * | 2019-08-28 | 2020-02-14 | 广东电网有限责任公司 | Intelligent safety monitoring system for electric power operation and monitoring method thereof |
-
2020
- 2020-04-17 CN CN202010306273.4A patent/CN111652046A/en active Pending
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180204111A1 (en) * | 2013-02-28 | 2018-07-19 | Z Advanced Computing, Inc. | System and Method for Extremely Efficient Image and Pattern Recognition and Artificial Intelligence Platform |
| US20180333083A1 (en) * | 2015-02-25 | 2018-11-22 | Leonardo Y. Orellano | Fall detection systems and methods |
| CN106295601A (en) * | 2016-08-18 | 2017-01-04 | 合肥工业大学 | A kind of Safe belt detection method of improvement |
| CN107680220A (en) * | 2017-09-28 | 2018-02-09 | 朱明增 | A kind of operation unlawful practice intelligent identification Method based on machine vision technique |
| CN107909034A (en) * | 2017-11-15 | 2018-04-13 | 清华大学深圳研究生院 | A kind of method for detecting human face, device and computer-readable recording medium |
| CN110119656A (en) * | 2018-02-07 | 2019-08-13 | 中国石油化工股份有限公司 | Intelligent monitor system and the scene monitoring method violating the regulations of operation field personnel violating the regulations |
| CN109543542A (en) * | 2018-10-24 | 2019-03-29 | 杭州叙简科技股份有限公司 | A kind of determination method whether particular place personnel dressing standardizes |
| CN109635697A (en) * | 2018-12-04 | 2019-04-16 | 国网浙江省电力有限公司电力科学研究院 | Electric operating personnel safety dressing detection method based on YOLOv3 target detection |
| CN109635758A (en) * | 2018-12-18 | 2019-04-16 | 武汉市蓝领英才科技有限公司 | Wisdom building site detection method is dressed based on the high altitude operation personnel safety band of video |
| CN109757791A (en) * | 2019-01-25 | 2019-05-17 | 武汉职业技术学院 | A kind of intelligent and safe school uniform device |
| CN110070055A (en) * | 2019-04-25 | 2019-07-30 | 南京邮电大学 | A kind of capital construction scene safety detecting system and method based on deep learning |
| CN110334619A (en) * | 2019-06-21 | 2019-10-15 | 上海翊视皓瞳信息科技有限公司 | A kind of secure supervision system and method based on wearable device |
| CN110309768A (en) * | 2019-06-28 | 2019-10-08 | 上海眼控科技股份有限公司 | The staff's detection method and equipment of car test station |
| 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 |
| CN110795989A (en) * | 2019-08-28 | 2020-02-14 | 广东电网有限责任公司 | Intelligent safety monitoring system for electric power operation and monitoring method thereof |
| CN110674772A (en) * | 2019-09-29 | 2020-01-10 | 国家电网有限公司技术学院分公司 | Intelligent safety control auxiliary system and method for electric power operation site |
| 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 |
Non-Patent Citations (4)
| Title |
|---|
| 周树道，贺宏兵著: "《现代气象雷达》", 31 December 2017 * |
| 施静，王天江，田波: "《学习记忆与机器学习实验原理》", 30 July 2019 * |
| 曹荣刚，张庆霞著: "《电磁轨道发射装置电磁环境及其测试技术Electromagnetic Environment and Parameter Measurement for Railgun》", 30 March 2019 * |
| 谢逸，张竞文，李韬，刘嘉；: "《基于视频监控的地铁施工不安全行为检测预警》", 《华中科技大学学报(自然科学版)》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112183264A (en) * | 2020-09-17 | 2021-01-05 | 国网天津静海供电有限公司 | Method for judging people lingering under crane boom based on spatial relationship learning |
| CN112183264B (en) * | 2020-09-17 | 2023-04-21 | 国网天津静海供电有限公司 | Method for judging someone remains under crane boom based on spatial relationship learning |
| CN112507945A (en) * | 2020-12-18 | 2021-03-16 | 南方电网电力科技股份有限公司 | Method and device for managing and controlling behavior of operator, electronic equipment and storage medium |
| CN114882442A (en) * | 2022-05-31 | 2022-08-09 | 广州信诚信息科技有限公司 | Personnel and equipment situation identification method based on electric power operation site |
| CN115601712A (en) * | 2022-12-15 | 2023-01-13 | 南京电力自动化设备三厂有限公司（Cn） | Image data processing method and system suitable for field safety measures |
| CN115601712B (en) * | 2022-12-15 | 2023-08-22 | 南京电力自动化设备三厂有限公司 | Image data processing method and system suitable for site safety measures |
| CN116958702A (en) * | 2023-08-01 | 2023-10-27 | 浙江钛比科技有限公司 | Hotel guard personnel wearing detection method and system based on edge artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111652046A (en) | Safe wearing detection method, equipment and system based on deep learning | |
| CN111598040A (en) | Construction worker identity identification and safety helmet wearing detection method and system | |
| CN112200081A (en) | Abnormal behavior identification method and device, electronic equipment and storage medium | |
| CN112990870A (en) | Patrol file generation method and device based on nuclear power equipment and computer equipment | |
| CN112507892A (en) | System, method and device for identifying and processing wearing of key personnel in special place based on deep learning, processor and storage medium thereof | |
| CN113780342A (en) | Intelligent detection method and device based on self-supervision pre-training and robot | |
| CN111091104A (en) | Target object protection detection method, device, equipment and storage medium | |
| CN111372042A (en) | Fault detection method and device, computer equipment and storage medium | |
| CN114049540A (en) | Method, device, equipment and medium for detecting marked image based on artificial intelligence | |
| CN117115823A (en) | Tamper identification method and device, computer equipment and storage medium | |
| CN115131826B (en) | Article detection and identification method, and network model training method and device | |
| CN107566826A (en) | The method of testing and device of grating image processor | |
| CN115311447A (en) | Pointer instrument indicating number identification method based on deep convolutional neural network | |
| CN114882442A (en) | Personnel and equipment situation identification method based on electric power operation site | |
| CN114550129A (en) | Machine learning model processing method and system based on data set | |
| CN114417965A (en) | Training method of image processing model, target detection method and related device | |
| CN112906581A (en) | Method, device, equipment and storage medium for detecting dressing standard of protective clothing | |
| CN111967419A (en) | Dam bank dangerous case prediction method and device, computer equipment and storage medium | |
| CN113033289A (en) | Safety helmet wearing inspection method, device and system based on DSSD algorithm | |
| CN115376275B (en) | Construction safety warning method and system based on image processing | |
| CN111225297A (en) | Broadband passive optical network port resource remediation method and system | |
| CN117253196B (en) | Video-based security risk monitoring method and device in steel industry | |
| CN114283413B (en) | Method and system for identifying digital instrument readings in inspection scene | |
| CN114463415A (en) | Social distance monitoring method, device, medium and equipment for preventing respiratory infection | |
| CN117809349A (en) | Method and device for detecting wearing of mask in financial website |
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: 20200911 |
|
| RJ01 | Rejection of invention patent application after publication |