CN111950457A - Oilfield Safety Production Image Recognition Method and System - Google Patents

Abstract

The invention discloses an image identification method and system for oil field safety production, which comprises the following steps: the method comprises the steps that a camera device shoots an oilfield field video in real time, and preprocessing and coding are carried out on pictures in the oilfield field video in batches to obtain video image data; sequentially carrying out dark channel filtering data enhancement design, receptive field design, activation function design, channel attention mechanism design, pyramid pooling design and unbalance training design on the video image data to obtain final recognition target same-size video image data; and decoding and frame sequencing the video image data of the final recognition target with the same size to obtain a recognition video, and outputting and marking the recognition target. Has the advantages that: the influence of weather on the blur of the taken picture is overcome; the receptive field is enlarged, more useful characteristic information is extracted, and the problem that the target is small and cannot be identified is solved. Finally, the recognition target is clearer, and the recognition rate is improved.

Description

Oilfield Safety Production Image Recognition Method and System

technical field

The invention relates to the technical field of oil safety, in particular to an image recognition method and system for oil field safety production.

Background technique

In the safety operation of oilfields, there are uncertain problems such as large operation mobility and overlapping business operations. On-site operations need to be equipped with corresponding protective equipment. The whole operation process involves unsafe factors of the environment, unsafe behavior of people, unsafe state of machines and management defects, which may bring immeasurable harm and loss.

For example, the wearing of safety helmets, work clothes, flames, smoke, intrusion of strangers, and workers' faces in oilfield safety operations all need to be strictly controlled. Therefore, in the oilfield operation area, some identification technology measures have appeared to solve the above problems.

Among them, the existing method of helmet wearing recognition is to use YOLOv3 to directly train and learn to detect the wearing, obtain the result by using the video stream detection of the previous frame image and predict the position and category information of the next frame, and then combine the detection frame and the prediction frame. The intersection and ratio correlation finally realizes the tracking. The detection of this method relies on the target detection algorithm and tracking of computer vision image processing technology. The core technology is to improve the detection and tracking effect by calculating the accuracy of the center point coordinates of the prediction frame. However, due to its own downsampling method, it is difficult to identify small targets and block problems. From the perspective of the algorithm itself, it is difficult to analyze small targets in oilfield production operation sites.

The identification of work clothes is to use the data based on the business scene to use the same method as the identification of helmets for training, testing, verification, and then to predict the category of scene image data and mark the area for detection and identification, and then the results are obtained for inventory and forensics. Due to the detection and identification of work clothes, the methods of the prior art are generally utilized in the same business scenario. Although this method can solve basic business problems through pedestrian detection under normal weather conditions, due to the diffuse reflection of the sensor in different weather conditions, the resolution distortion of image data and video results in a large number of false identification problems.

Flame recognition is to collect evidence through the same algorithm video, image marking preprocessing, flame recognition and smoke to locate the smoke and flame occurrence area and timely alarm, at the same time store the relevant results forensics and store them in the database for analysis and other services. The traditional flame and smoke detection also uses the prior art method to identify, and there is a problem of false alarms in foggy weather.

To sum up, the above problems cannot be fully solved in oilfield safety operations, and there are still great potential safety hazards.

SUMMARY OF THE INVENTION

In view of the above problems, the present invention provides an image recognition method for oilfield safety production. The video images of the oil production site are acquired, and the design is enhanced by dark channel filtering data, which eliminates weather interference, makes the image more accurate and clearer, and combines with the receptive field. Design, activation function design, channel attention mechanism design, activation function design, channel attention mechanism design, pyramid pooling design and unbalanced training design to identify objects and extract features quickly and accurately. It is convenient to supervise the situation on the production site.

In order to achieve the above object, the concrete technical scheme adopted in the present invention is as follows:

An oilfield safety production image recognition method, the key technology of which is: the specific steps are:

S1: The camera equipment captures the field video of the oil field in real time, and preprocesses and encodes the pictures in the field video of the oil field in batches to obtain the video image data;

S2: According to the weather period, perform a dark channel filtering data enhancement design on the video image data to obtain enhanced video image data;

S3: Design the receptive field for the enhanced video image data, and obtain the video image data with a certain field of view;

S4: Determine the recognition features, set the recognition thresholds of the recognition features correspondingly, design the activation function and the channel attention mechanism for the video image data of the field of view determination range, and extract the final recognition target in the video image data of the field of view determination range;

S5: Perform pyramid pooling design and unbalanced training design for each final recognition target in the video image data of the visual field-determined range to obtain video image data of the same size as the final recognition target;

S6: After decoding the video image data of the same size of the final recognition target and sorting the frames, the recognition video is obtained, and the recognition target is output and marked.

每张图片的深度值、宽度值、高度值。其中每一张输入图像的像素总数|w|、输入图像的像素方差

可经中间计算得到，该数据计算属于现有技术，不做赘述。Further, in order to carry out image recognition design processing, the video image data includes at least the arrangement order of all pictures in the video, the original pixel value of each frame of picture in the video, the length value and width value of each picture, each in the video image data. The total number of pixels in an input image |w|, the pixel variance of the input image

The depth, width, and height values of each image. The total number of pixels in each input image |w|, the pixel variance of the input image

It can be obtained through intermediate calculation, and the data calculation belongs to the prior art and will not be described in detail.

A further technical solution is: the enhanced video image data described in step S2 is the image data corresponding to the pixel value obtained from the enhancement design, and the calculation function of the image pixel value qi _outputted after the enhancement design is:

q _i = _ak I _i +b _k ;

q _i is the pixel value of the image in the output enhanced video image data; I _i is the original pixel value of the picture;

k and i are the pixel index values, a _k , b _k are the coefficients of the function at time k of the center coordinate of the current picture; the value of i is 0-255;

取值最小时对应的取值；P_i为输入图像的向量，ε为调整图像模糊程度参数，取值范围为0-2；∑iew_k为输入损失函数；系数a_k的计算公式为：Among them, the coefficient b _k is a linear value function

The corresponding value when the value is the smallest; P _i is the vector of the input image, ε is the parameter for adjusting the blur degree of the image, and the value range is 0-2; ∑iew _k is the input loss function; the calculation formula of the coefficient a _k is:

为输入图像的像素方差。|w| is the total number of pixels of each input image;

is the pixel variance of the input image.

The above calculation method is used to obtain constant coefficients to adjust the image to solve the problem that the system recognition rate is low due to the weather. Then the image data labeling is mainly to normalize the images collected by the image sensor using the RGB three-channel of x*y pixels, and then the video is framed and labeled with the labeling tool. In the early training process, we use semi-automatic training recognition and then model The pseudo-labels are used to analyze the residuals of the loss function and then enhance the labeling training on data that is too volatile.

A further technical solution is to design a receptive field in order to extract more feature information. Wherein, after the receptive field design in step S3, the picture size O of the video image data of the visual field determination range is obtained as:

Among them, i' is the length value and width value of the picture in the video, the two are equal; s is the step size of the convolution kernel; p is the pixel value of the corresponding field of view area when increasing the perceived field of view; k' is the size of the convolution kernel; is an odd number between 3-9; d is the introduced hyperparameter; (d-1) is the number of spaces inserted when increasing the perceived field of view.

To further describe, the identification features are safety helmet wearing identification features, work clothes identification features, flame identification features, stranger intrusion identification features, and face identification features.

Described further, the activation function design adopts Mish activation function, specifically;

Mish=x*tanh(ln(1+e ^x ))

ln is the logarithm, e is a natural constant, x is the video image data of the visual field to determine the range; tanh is the high-quality feature image data after convolution filtering;

Through this activation function, the useful information utilization of the model is improved to better generalize and accurately output high-quality feature information.

To further describe, the channel attention mechanism design function is:

input: c×w×h

globalpooling:c×1×1

FC:

FC: c×1×1

sigmoid: c×1×1

input=tanh, c w h are the depth, width and height of the input image before the channel attention mechanism is designed; globalpooling is the global pooling; c is the output image depth after the channel attention mechanism is designed; sigmoid is a sub-function of the Mish activation function;

FC is the output prediction value for the fully connected layer, and the output prediction value includes the predicted target category and the probability of each predicted target category; according to the set recognition threshold, the final recognition target in the video image data of the visual field determination range is extracted.

Appropriately increase the amount of calculation and use the feature information module design more efficiently and accurately. The attention mechanism in the field of image recognition and target detection generally focuses on soft attention. The design usually expresses features for regional features. Soft attention pays more attention to regions or Channels enable deterministic attentional representation through the network.

SEblock is mainly composed of Squeeze and Excitation mapping. Since convolution is only a local space calculation problem, it is difficult to obtain global channel information. In the analysis network, the importance information collected by the global information through the receptive field requires a Squeeze feature encoding method to encode it as a global feature. However, global average pooling is used.

Sequeeze is a method of global feature description to extract the relationship between channels to express nonlinear features that are not mutually exclusive between channels. At the same time, in order to reduce the complexity of the model and improve the generalization ability of the model, the bottleneck of the fully connected layer is used. One FC layer of the structure reduces dimensionality.

Excitation is that the coefficient of this dimension is restored to the original dimension by using mish activation, and finally the gating multiplication is realized by using sigmoid activation method for each channel.

The channel attention of the above SEblock module is mainly to solve the detection of the occlusion problem of our target in this project. Combined with the information context of the surrounding area of the channel information, the accuracy of the model is better improved.

To further describe, the pyramid pooling design uses the SPP Block pyramid pooling module to perform scale training on each batch of pictures, and outputs pictures with the same size.

Through the above design, we only need to perform fine-tuning model update when the equipment is transformed for oilfield digital images, so as to realize multi-scale training and recognition, and solve the sensitivity of the model to different sizes, ensuring the model output. The prediction tensor size is uniform.

An oilfield safety production image recognition system, the key technology of which is: comprising an acquisition module, an algorithm module, and a cognitive computing module, wherein the acquisition module is connected with the algorithm module, and the algorithm module is connected with the cognitive computing module; the acquisition module A camera device, a data screening unit, and an enhancement design unit are provided inside; the algorithm module is provided with a receptive field design unit, an activation function design unit, a channel attention mechanism design unit, a pyramid pooling design unit, and an unbalanced training design unit; The cognitive computing module includes a microservice unit.

Among them, the acquisition module is used for acquiring video data, screening, enhancing, marking sorting, batching, etc.

The store algorithm system is set up in the algorithm module, such as face recognition algorithm, CAFENet algorithm, encryption container, etc., which are used for receptive field design unit, activation function design unit, channel attention mechanism design unit, pyramid pooling design unit, unbalanced training design unit.

The cognitive computing module includes micro-service units, which mainly use distributed micro-service computing to realize dynamic resource scheduling through HTTP request density, and realize functions such as face recognition.

Compared with the prior art, the beneficial effects of the present invention:

1. The dark channel filter data enhancement design is adopted. When a large number of video data are collected under different weather conditions, guided filtering is required to enhance the data of the dark channel video image, which effectively overcomes the image distortion caused by weather changes and weather conditions. The dark channel filtering data enhancement design is adopted, and constant coefficients are obtained to adjust the image to solve the problem that the system recognition rate is low due to the weather.

2. Through the design of the receptive field, the receptive field can be expanded and more useful feature information can be extracted.

3. Through activation function design and channel attention mechanism design, higher quality and more accurate recognition targets are obtained. The problem that the target of the safety helmet is small and cannot be identified in the prior art is solved.

4. Through pyramid pooling design and unbalanced training design, the same batch of output data is consistent, and through unbalanced weighting design, the final recognition target is clearer and the recognition rate is improved.

Description of drawings

Fig. 1 is the identification method flow chart of the present invention;

Fig. 2 is the arbitrary frame picture in the video of oil safety production site;

Fig. 3 is the recognition effect diagram to Fig. 2;

Figure 4 is a block diagram of the identification system.

Note: Since it involves image recognition, Figures 2 and 3 are submitted in color.

Detailed ways

The specific embodiments and working principles of the present invention will be further described in detail below with reference to the accompanying drawings.

It can be seen from Figure 1 that a method for image recognition of oilfield safety production includes the following steps:

S1: The camera equipment captures the field video of the oil field in real time, and preprocesses and encodes the pictures in the field video of the oil field in batches to obtain the video image data;

In this embodiment, the video image data includes the arrangement order of all pictures in the video, the original pixel value of each frame of pictures in the video, the length and width values of each picture, and the pixels of each input image in the video image data. The total number |w|, the pixel variance σ _k ² of the input image, the depth value, width value, and height value of each image.

S2: According to the weather period, perform a dark channel filtering data enhancement design on the video image data to obtain enhanced video image data;

The enhanced video image data described in step S2 is the image data corresponding to the pixel value obtained from the enhanced design, and the calculation function of the image pixel value qi _outputted after the enhanced design is:

q _i = _ak I _i +b _k ;

q _i is the pixel value of the image in the output enhanced video image data; I _i is the original pixel value of the picture;

k and i are the pixel index values, a _k , b _k are the coefficients of the function at time k of the center coordinate of the current picture; the value of i is 0-255;

取值最小时对应的取值；P_i为输入图像的向量；Among them, the coefficient b _k is a linear value function

The value corresponding to the smallest value; P _i is the vector of the input image;

ε is a parameter for adjusting the blur degree of the image, and in this embodiment, the value range of ε is 2.

∑iew _k is the input loss function;

The formula for calculating the coefficient a _k is:

为输入图像的像素方差。|w| is the total number of pixels of each input image;

is the pixel variance of the input image.

In this embodiment, the enhanced video image data can be obtained according to the pixel value _qi of the output image.

The above calculation method is used to obtain constant coefficients to adjust the image to solve the problem that the system recognition rate is low due to the weather. Then the image data is marked mainly by normalizing the images collected by the image sensor with RGB three channels of 906*720 pixels, and then the video is framed and marked with the labeling tool. In the early stage of the training process, we use semi-automatic training recognition and then model The pseudo-labels are used to analyze the residuals of the loss function and then enhance the labeling training on data that is too volatile.

S3: Design the receptive field for the enhanced video image data, and obtain the video image data with a certain field of view;

In the present embodiment, the calculation formula for obtaining the picture size O of the video image data of the visual field determination range after designing the receptive field in S3 is:

Among them, i' is the length value and width value of the picture in the video, the two are equal; s is the step size of the convolution kernel; p is the pixel value of the corresponding field of view area when increasing the perceived field of view;

k' is the size of the convolution kernel; it is an odd number between 3-9;

In this embodiment, during the convolution calculation process, in the initial stage, the convolution kernel is set to 9 and 7, and as the convolution proceeds, the size of the convolution kernel is finally set to 3.

d is the introduced hyperparameter; (d-1) is the number of spaces inserted when increasing the perception field.

S4: Determine the recognition features, set the recognition thresholds of the recognition features correspondingly, design the activation function and the channel attention mechanism for the video image data of the field of view determination range, and extract the final recognition target in the video image data of the field of view determination range;

In this embodiment, the identification features include safety helmet wearing identification features, work clothes identification features, flame identification features, stranger intrusion identification features, and face identification features.

In this embodiment, the activation function design adopts the Mish activation function, specifically:

Mish=x*tanh(ln(1+e ^x ))

ln is the logarithm, e is a natural constant, and x is the video image data of the visual field to determine the range; tanh is the high-quality feature image data after convolution filtering; the design function of the channel attention mechanism is:

input: c×w×h

globalpooling:c×1×1

FC:

FC: c×1×1

sigmoid: c×1×1

input=tanh, c w h are the depth, width and height of the input image before the channel attention mechanism is designed; globalpooling is the global pooling; c is the output image depth after the channel attention mechanism is designed; sigmoid is a sub-function of the Mish activation function;

FC is the output prediction value for the fully connected layer, and the output prediction value includes the predicted target category and the probability of each predicted target category; according to the set recognition threshold, the final recognition target in the video image data of the visual field determination range is extracted.

S5: Perform pyramid pooling design and unbalanced training design for each final recognition target in the video image data of the visual field-determined range to obtain video image data of the same size as the final recognition target;

The pyramid pooling design uses the SPP Block pyramid pooling module to perform scale training on each batch of pictures, and outputs pictures with the same size.

S6: After decoding the video image data of the same size of the final recognition target and sorting the frames, the recognition video is obtained, and the recognition target is output and marked.

An oilfield safety production image recognition system, as can be seen from Figure 4, includes an acquisition module 1, an algorithm module 2, and a cognitive computing module 3. The acquisition module 1 is connected to the algorithm module 2, and the algorithm module 2 is connected to the cognitive computing module 3. The computing module 3 is connected; the acquisition module 1 is provided with a camera device, a data screening unit, and an enhancement design unit; the algorithm module 2 is provided with a receptive field design unit, an activation function design unit, a channel attention mechanism design unit, and a pyramid. Pooling design unit and unbalanced training design unit; the cognitive computing module 3 includes a micro-service unit.

In this embodiment, it can be seen from the comparison of FIG. 2 and FIG. 3 that the mark identification target is frame-selected and marked by using wire frames of green, red, purple, blue and other colors.

After outputting the identification target, output the identification label. In this embodiment, the identification label includes: normal (normal state), nohat (no helmet), nocoverall (no work clothes), flame (fire), smoke (smoke). ).

In this embodiment, it can be seen from FIG. 3 that a green wire frame is used to frame the identification target without a helmet; a blue wire frame is used to frame the identification target wearing a helmet;

In this embodiment, it can be seen in conjunction with FIG. 3 that a purple line frame is used to select a recognition target wearing work clothes; a red wire frame is used to select a recognition target that does not wear work clothes.

It should be noted that the above descriptions are not intended to limit the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by those of ordinary skill in the art within the scope of the present invention, It should also belong to the protection scope of the present invention.

Claims (8)

1. An image identification method for oil field safety production is characterized by comprising the following specific steps:

s1: the method comprises the steps that a camera device shoots an oilfield field video in real time, and preprocessing and coding are carried out on pictures in the oilfield field video in batches to obtain video image data;

s2: according to the weather time interval condition, dark channel filtering data enhancement design is carried out on video image data to obtain enhanced video image data;

s3: carrying out receptive field design on the enhanced video image data to obtain video image data with a visual field determining range;

s4: determining identification characteristics, correspondingly setting identification thresholds of the identification characteristics, performing activation function design and channel attention mechanism design on the video image data in the visual field determination range, and extracting a final identification target in the video image data in the visual field determination range;

s5: carrying out pyramid pooling design and unbalanced training design on each final recognition target in the video image data of the visual field determining range to obtain the video image data of the final recognition target in the same size;

s6: and decoding and frame sequencing the video image data of the final recognition target with the same size to obtain a recognition video, and outputting and marking the recognition target.

2. The image recognition method for oilfield safety production according to claim 1, characterized in that: the video image data at least comprises the arrangement sequence of all pictures in the video, the original pixel value of each frame of picture in the video, the length value and the width value of each picture, the total pixel number | w | of each input image in the video image data, and the pixel variance of the input image

Depth value, width value, height value of each picture.

3. The image recognition method for oilfield safety production according to claim 2, characterized in that: the enhanced video image data in step S2 is image data obtained by solving the enhanced design to a pixel value, and the image pixel value q output after the enhanced design_iThe calculation function of (a) is:

q_i＝a_kI_i+b_k；

q_ipixel values for an image in the output enhanced video image data; i is_iIs the original pixel value of the picture;

k and i are pixel index values, a_k,b_kThe coefficient of the function at the moment k is the central coordinate of the current picture; the value of i is 0 to 255;

wherein the coefficient b_kIs a linear cost function

The value corresponding to the minimum value; p_iThe vector of the input image is used for adjusting the image blurring degree parameter, and the value range is 0-2; sigma iew_kIs an input loss function; coefficient a_kThe calculation formula of (2) is as follows:

the | w | is the total number of pixels of each input image;

is the pixel variance of the input image.

4. The image recognition method for oilfield safety production according to claim 2, characterized in that: the picture size O of the video image data of the view determination range obtained after the view design in step S3 is:

wherein i' is the length value and the width value of the picture in the video, and the length value and the width value are equal; s is the convolution kernel step length; p is the filling pixel value of the corresponding visual field area when the visual field is increased; k' is the size of the convolution kernel; is an odd number between 3 and 9; d is an introduced hyper-parameter; (d-1) the number of empty cells inserted for increasing the perceived visual field.

5. The image recognition method for oilfield safety production according to claim 2, characterized in that: the identification features are safety helmet wearing identification features, worker clothes identification features, flame identification features, stranger intrusion identification features and face identification features.

6. The image recognition method for oilfield safety production according to claim 5, wherein: the activation function is designed by adopting a Mish activation function, and specifically, the Mish activation function is designed;

Mish＝x*tanh(ln(1+e^x))

ln is logarithm, e is natural constant, x is visual field determining range video image data; tanh is high-quality characteristic image data after convolution filtering;

the channel attention mechanism design function is as follows: :

input:c×w×h

globalpooling:c×1×1

FC:

FC:c×1×1

sigmoid:c×1×1

input is tan h, and c w h is the depth, width and height of the input image before the channel attention mechanism is designed; globalpopooling is global pooling; c, outputting the image depth after the channel attention mechanism is designed; sigmoid is a subfunction of the Mish activation function;

FC is the output prediction value of the full connection layer, and the output prediction value comprises the prediction target category and the probability of each prediction target category; and extracting a final recognition target in the visual field determination range video image data according to a set recognition threshold.

7. The image recognition method for oilfield safety production according to claim 6, wherein: and the pyramid pooling design adopts an SPP Block pyramid pooling module to perform scale training on each batch of pictures and outputs the pictures with consistent sizes.

8. The oil field safety production image recognition system is characterized in that: the cognitive computing system comprises an acquisition module (1), an algorithm module (2) and a cognitive computing module (3), wherein the acquisition module (1) is connected with the algorithm module (2), and the algorithm module (2) is connected with the cognitive computing module (3);

the acquisition module (1) is internally provided with a camera device, a data screening unit and an enhancement design unit;

a receptive field design unit, an activation function design unit, a channel attention mechanism design unit, a pyramid pooling design unit and an unbalance training design unit are arranged in the algorithm module (2);

the cognitive computation module (3) comprises a micro-service unit.

Images