CN118097553A - Pedestrian number determining method and device and related equipment - Google Patents

Abstract

The present invention provides a method, device and related equipment for determining the number of pedestrians, which relates to the field of artificial intelligence technology. The method includes: determining the number of pedestrians in a first pixel area in a first image acquired by a collection device; determining the pedestrian number interval in the second pixel area based on a second pixel area of the first image; determining the number of pedestrians in the first image based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area corresponding to the pedestrian number interval; the distance between the first position area mapped by the first pixel area and the collection device is less than the distance between the second position area mapped by the second pixel area and the collection device. Through the method for determining the number of pedestrians that combines quantitative and qualitative methods, pedestrians can be detected and identified more accurately and efficiently, and the accuracy of statistical results is improved.

Description

A method, device and related equipment for determining the number of pedestrians

Technical Field

The present invention relates to the field of artificial intelligence technology, and in particular to a method and device for determining the number of pedestrians and related equipment.

Background technique

With the development of urbanization, urban traffic is becoming increasingly busy, and the corresponding demand for public safety is also increasing. In order to better improve the safety of public places (such as supermarkets, subway stations, bus stations and railway stations, etc.), pedestrians can be counted to facilitate the formulation of emergency measures for emergencies.

In the scene of pedestrians queuing, the camera is usually set in front of or behind the queue. When there are many people in the queue and the queue is long, the pedestrians far away from the camera become very small and dense in the picture captured by the camera, and the occlusion will be very serious, resulting in large errors in counting the number of pedestrians.

Summary of the invention

The embodiments of the present invention provide a method, an apparatus and related equipment for determining the number of pedestrians, so as to solve the problem of large error in determining the number of pedestrians in the prior art.

To solve the above technical problems, the present invention is achieved as follows:

In a first aspect, an embodiment of the present invention provides a method for determining the number of pedestrians, the method comprising:

Determine, based on a first image acquired by an acquisition device, the number of pedestrians in a first pixel area in the first image;

Determining a pedestrian quantity interval of a second pixel area based on a second pixel area of the first image;

Determining the number of pedestrians in the first image based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area corresponding to the pedestrian number interval;

The distance between the first position area mapped by the first pixel area and the acquisition device is smaller than the distance between the second position area mapped by the second pixel area and the acquisition device.

Optionally, determining the pedestrian quantity interval of the second pixel area based on the second pixel area of the first image includes:

According to the number of first target pixels, the number of second target pixels, and the number of pedestrians in the first pixel area, the pedestrian number interval of the second pixel area is determined, the first target pixel is a pixel having a target feature in the first pixel area, the second target pixel is a pixel having the target feature in the second pixel area, and the target feature is a feature that characterizes pedestrians.

Optionally, determining the pedestrian quantity interval of the second pixel area based on the second pixel area of the first image includes:

The second pixel area of the first image is detected according to an image classification algorithm, and a pedestrian quantity interval corresponding to the second pixel area is determined from a plurality of preset pedestrian quantity intervals.

Optionally, determining the number of pedestrians in a first pixel area in the first image based on the first image acquired by the acquisition device includes:

Detect the first image according to the Yolo object detection algorithm to obtain the number of pedestrians in the first pixel area;

A pixel region other than the first pixel region in the first image is determined as the second pixel region.

Optionally, the first pixel area is a pixel area marked by the Yolo object detection algorithm in the first image.

Optionally, the determining the number of pedestrians in the first image based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area corresponding to the pedestrian number interval includes:

Obtaining a first predicted number of pedestrians in the first image according to the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area;

When the difference between the first predicted number and the second predicted number is less than or equal to a preset threshold, determining the first predicted number as the number of pedestrians in the first image;

The second predicted number is the number of pedestrians determined based on the historical number of pedestrians and external factors, and the external factors include weather factors and time factors when the acquisition device acquires the first image.

In a second aspect, an embodiment of the present invention provides a device for determining the number of pedestrians, the device comprising:

A first determination module, configured to determine the number of pedestrians in a first pixel area in a first image acquired by an acquisition device based on the first image;

A second determination module, configured to determine a pedestrian quantity interval of a second pixel area based on a second pixel area of the first image;

a third determination module, configured to determine the number of pedestrians in the first image based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area corresponding to the pedestrian number interval;

The distance between the first position area mapped by the first pixel area and the acquisition device is smaller than the distance between the second position area mapped by the second pixel area and the acquisition device.

In a third aspect, an embodiment of the present invention provides an electronic device, including a transceiver and a processor.

The processor is used to determine the number of pedestrians in a first pixel area in the first image based on the first image acquired by the acquisition device;

The processor is further configured to determine a pedestrian quantity interval of the second pixel area based on the second pixel area of the first image;

The processor is further configured to determine the number of pedestrians in the first image based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area corresponding to the pedestrian number interval;

The distance between the first position area mapped by the first pixel area and the acquisition device is smaller than the distance between the second position area mapped by the second pixel area and the acquisition device.

In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method for determining the number of pedestrians as described in the first aspect are implemented.

In a fifth aspect, an embodiment of the present invention provides a computer program product, comprising computer instructions, which, when executed by a processor, implement the steps of the method for determining the number of pedestrians as described in the first aspect.

In an embodiment of the present invention, a quantitative detection is first performed on pedestrians in the first image to determine the number of pedestrians in the first pixel area in the first image, and the pixel area in the first image other than the first pixel area is determined as the second pixel area; then a qualitative detection is performed on pedestrians in the second pixel area in the first image to determine the number interval of pedestrians in the second pixel area, thereby reducing the statistical error caused by missing the second pixel area that cannot be recognized by the object detection algorithm. By combining quantitative and qualitative methods to determine the number of pedestrians, pedestrians can be detected and identified more accurately and efficiently, thereby improving the accuracy of statistical results. This provides better technical support for the fields of public safety, intelligent transportation, etc.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

FIG1 is a flow chart of a method for determining the number of pedestrians provided by an embodiment of the present invention;

FIG2 is a second flowchart of a method for determining the number of pedestrians provided by an embodiment of the present invention;

FIG3 is a schematic diagram of the structure of a device for determining the number of pedestrians provided by an embodiment of the present invention;

FIG. 4 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention.

Detailed ways

The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

Referring to FIG. 1 , FIG. 1 is a flow chart of a method for determining the number of pedestrians provided by an embodiment of the present invention. As shown in FIG. 1 , the method includes the following steps:

Step 101: determining the number of pedestrians in a first pixel area in a first image based on a first image acquired by a collection device;

The acquisition device may be a visual acquisition device such as a video camera, and the acquisition device may be set at the entrance and exit of a public place (such as a supermarket, a subway station, a bus station, and a railway station, etc.) to facilitate pedestrian counting. The first image may be any frame image in the video stream data captured by the acquisition device.

In some optional implementations, determining the number of pedestrians in a first pixel area in the first image based on the first image acquired by the acquisition device includes:

Detect the first image according to the Yolo object detection algorithm to obtain the number of pedestrians in the first pixel area;

A pixel region other than the first pixel region in the first image is determined as the second pixel region.

In this implementation, a Yolo (You Only Look Once)-based object detection algorithm may be used to quantitatively detect the first image to determine the number of pedestrians in the first pixel region of the first image. The training process of the Yolo-based object detection algorithm model may be described as follows:

First, data collection: Use collection equipment (taking surveillance cameras as an example) to collect image data. The camera resolution can include 1080p, 720p, 480p and other formats. At the same time, you can also collect some historical image data of different sizes to expand the training data set. These image data are screened to find image data that meets the requirements of pedestrian detection, and then converted into pictures of a unified format as sample images.

Then, data annotation: use annotation tools to annotate pedestrians in sample images. To avoid occlusion and ensure that pedestrians are easy to identify, the head and shoulders of pedestrians can be annotated. The annotation box can be a rectangular annotation box consisting of two points.

Training: Training is done on the official model, which includes C3 module, Conv module, SPPF module and ConCat module. After training, the object detection algorithm model based on Yolo is obtained.

In this way, the first image acquired by the acquisition device is input into the object detection algorithm model based on Yolo, and the number of pedestrians in the first pixel area can be obtained. In the embodiment of the present invention, it is considered that the object detection algorithm model based on Yolo has limited recognition ability in the process of pedestrian detection. For example, in the scene of pedestrians queuing, the camera is generally set in front of or behind the queue. When the number of people in the queue is large and the queue is long, the pedestrians far away from the camera become very small and dense in the first image collected by the camera, and the occlusion will also be very serious, which makes it difficult for the object detection algorithm model based on Yolo to detect and count the pedestrians in this area. Therefore, the object detection algorithm model based on Yolo can be used to make predictions on the test set; then the obtained test results are analyzed to determine the area that the object detection algorithm model based on Yolo cannot recognize. In other words, the first pixel area is the pixel area marked by the Yolo object detection algorithm in the first image, that is, the first pixel area is the area where pedestrians can be recognized in the first image; the pixel area in the first image other than the first pixel area is determined as the second pixel area.

Since the second pixel region may also contain pedestrians, the second pixel region may be further qualitatively detected in step 102 to reduce errors when counting the number of pedestrians.

Step 102: determining a pedestrian quantity interval of a second pixel area based on a second pixel area of the first image;

In one example, pedestrians in a first image are first quantitatively detected to determine the number of pedestrians in a first pixel area in the first image, and a pixel area other than the first pixel area in the first image is determined as a second pixel area; then, pedestrians in a second pixel area in the first image are qualitatively detected to determine an interval of the number of pedestrians in the second pixel area, thereby reducing statistical errors caused by missing the second pixel area that cannot be recognized by the object detection algorithm.

In some optional implementations, determining the pedestrian quantity interval of the second pixel area based on the second pixel area of the first image includes:

The second pixel area of the first image is detected according to an image classification algorithm, and a pedestrian quantity interval corresponding to the second pixel area is determined from a plurality of preset pedestrian quantity intervals.

In this embodiment, a qualitative analysis of the unrecognizable area can be performed based on an image classification algorithm, that is, a qualitative detection is performed on the second pixel area of the first image to determine the pedestrian number interval corresponding to the second pixel area in the first image. Among them, the second pixel area of the first image is detected according to the image classification algorithm, and the second pixel area can be qualitatively detected based on a deep residual network (ResNet) model. The training process of the ResNet model can be described as follows:

First, data collection: Pedestrian detection is performed on sample images based on the Yolo object detection algorithm model. The second pixel area in the sample image that cannot be detected by the object detection algorithm model is extracted to create a training data set for the ResNet model. According to the number of pedestrians, it is divided into multiple preset pedestrian number intervals from small to large: no people, few people (1 to 5 people), some people (about 5-10 people), crowded (more people), full, etc. The obtained training data set is divided into a training set and a test set.

Among them, the training data set needs to be preprocessed before inputting into the model to ensure that the input format meets the model requirements. The preprocessing steps may include adjusting the image size, normalizing pixel values, data enhancement (such as random cropping, rotation), etc., in order to improve the generalization ability and processing efficiency of the model.

Then, training is performed: a deep residual network (ResNet) model is used to perform qualitative detection on the second pixel area. ResNet is a deep convolutional neural network, which is famous for introducing residual learning units to solve the problem of difficult training of deep networks. First, according to the complexity of the task and the available computing resources, ResNet architectures of different depths can be selected, such as ResNet18, ResNet34, ResNet50, etc. The model is built using the selected ResNet architecture, and the output of the last layer of the network is adjusted according to the number of categories of the actual classification task. If a pre-trained model is used, the parameters of other layers except the last layer can be kept unchanged to utilize the existing feature extraction capabilities; the last layer (usually a fully connected layer) needs to be retrained for the new classification task. Then, the loss function and optimizer are defined, and the ResNet model is trained through multiple iterations. In each iteration, the ResNet model predicts the output of the training data, calculates the loss by comparing the predicted results with the true labels, and updates the weight parameters through the back-propagation algorithm to reduce the loss. During the training process, it is necessary to monitor the performance on the training set and the test set to adjust the training parameters or perform early stopping to prevent overfitting.

After training is completed, an independent test data set is used to evaluate the performance of the ResNet model. Common evaluation indicators include accuracy, precision, recall, etc. This step is an important means to test the generalization ability of the ResNet model. Finally, the trained ResNet model is deployed to actual applications for image classification tasks. When encountering new images, the ResNet model can identify the category of the image based on the learned features. The whole process is iterative and cyclical. Based on the evaluation results and feedback from actual applications, it may be necessary to return to the steps of data preparation, ResNet model adjustment, or retraining to continuously improve the performance of the ResNet model to meet actual needs.

In this way, the first image acquired by the acquisition device is first input into the object detection algorithm model based on Yolo, and the number of pedestrians in the first pixel area can be obtained. At the same time, the second pixel area in the first image that cannot be recognized by the object detection algorithm model is determined, and the second pixel area of the first image is further qualitatively detected by the image classification algorithm model based on ResNet to determine the number of pedestrians in the second pixel area. The pedestrian counting problem is converted into a classification problem, which reduces the difficulty of determining the number of pedestrians in the first image. Finally, according to step 103, the number of pedestrians in the first image is determined based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area corresponding to the pedestrian number interval. The image classification algorithm and the object detection algorithm are combined. Both algorithms use convolutional neural networks (CNNs), attention mechanisms, etc. to extract the features of pedestrians, reduce the statistical errors caused by omitting the second pixel area that cannot be recognized by the object detection algorithm, and improve the accuracy of the detection results.

Step 103: determining the number of pedestrians in the first image based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area corresponding to the pedestrian number interval;

The distance between the first position area mapped by the first pixel area and the acquisition device is smaller than the distance between the second position area mapped by the second pixel area and the acquisition device.

Taking the scenario of pedestrians queuing, where the camera is set in front of the queue as an example, the first position area mapped by the first pixel area is the front area of the pedestrian queue close to the camera, and the second position area mapped by the second pixel area is the rear area of the pedestrian queue away from the camera, that is, the distance between the first position area mapped by the first pixel area and the camera is smaller than the distance between the second position area mapped by the second pixel area and the camera.

In an embodiment of the present invention, a quantitative detection is first performed on pedestrians in the first image to determine the number of pedestrians in the first pixel area in the first image, and the pixel area in the first image other than the first pixel area is determined as the second pixel area; then a qualitative detection is performed on pedestrians in the second pixel area in the first image to determine the number interval of pedestrians in the second pixel area, thereby reducing the statistical error caused by missing the second pixel area that cannot be recognized by the object detection algorithm. By combining quantitative and qualitative methods to determine the number of pedestrians, pedestrians can be detected and identified more accurately and efficiently, thereby improving the accuracy of statistical results. This provides better technical support for the fields of public safety, intelligent transportation, etc.

In some optional embodiments, determining the pedestrian quantity interval of the second pixel area based on the second pixel area of the first image includes:

According to the number of first target pixels, the number of second target pixels, and the number of pedestrians in the first pixel area, the pedestrian number interval of the second pixel area is determined, the first target pixel is a pixel having a target feature in the first pixel area, the second target pixel is a pixel having the target feature in the second pixel area, and the target feature is a feature that characterizes pedestrians.

The features that characterize pedestrians, namely the target features, can be the features of the entire pedestrian or the features of a specific part of the pedestrian, such as the features of the pedestrian's head, eyes, shoulders or feet. In this example, the number of pedestrians in the first pixel area can be a numerical value quantitatively output by the object detection algorithm; then, based on the ratio between the number of pixels with target features located in the first pixel area (i.e., the first target pixels) and the number of pixels with target features located in the second pixel area (i.e., the second target pixels) in the first image, the pedestrian number interval in the second pixel area is determined, thereby reducing the statistical error caused by missing the second pixel area that cannot be recognized by the object detection algorithm and improving the accuracy of the statistical results.

In one embodiment, in a scenario where pedestrians are queuing, the ratio between the number of the first target pixels and the number of the second target pixels can be converted into a ratio between the length corresponding to the first target pixel in the first image and the length corresponding to the second target pixel in the first image.

For example, in the first image, pedestrians line up from position A to position B, position A is a position close to the camera, and position B is a position far from the camera. The first image is input into the Yolo-based object detection algorithm model, and the number of pedestrians in the first pixel area in the first image is determined to be R1. Among them, the first pixel area is the pixel area marked by the Yolo object detection algorithm in the first image, that is, the first pixel area is the area where pedestrians can be identified in the first image. In this way, the first pixel area can be the area corresponding to position A to position C in the first image, and position C is the position between position A and position B. Then, the pixel area in the first image except the first pixel area is determined as the second pixel area. In other words, the area corresponding to position C to position B in the first image is determined as the second pixel area.

Furthermore, the length AC from position A to position C and the length CB from position C to position B are determined. Thus, according to AC, CB and R1, the pedestrian number interval of the second pixel area can be determined as R2, R2 = (CB × R1) / AC. Thus, the number of pedestrians in the first image can be determined. By combining quantitative and qualitative methods for determining the number of pedestrians, pedestrians can be detected and identified more accurately and efficiently, thereby improving the accuracy of the statistical results.

In another embodiment, in a scene where pedestrians are scattered, the ratio between the number of the first target pixels and the number of the second target pixels can be converted into a ratio between an area corresponding to the first target pixel in the first image and an area corresponding to the second target pixel in the first image.

For example, pedestrians are scattered in various locations in the first image. The first image is input into the Yolo-based object detection algorithm model, and the number of pedestrians in the first pixel area in the first image is determined to be R1. The first pixel area is the pixel area marked by the Yolo object detection algorithm in the first image, that is, the first pixel area is the area where pedestrians can be identified in the first image. In this way, the first pixel area can be area A in the first image. Then, the pixel area other than the first pixel area in the first image is determined as the second pixel area. In other words, area B other than area A in the first image is determined as the second pixel area.

Further, the area of pixels with pedestrian features in region A is determined to be M1, and the area of pixels with pedestrian features in region B is determined to be M2. Thus, based on M1, M2 and R1, the pedestrian number interval of the second pixel region can be determined to be R2, R2 = (M2 × R1) / M1. Thus, the number of pedestrians in the first image can be determined, and the pedestrian number determination method combining quantitative and qualitative methods can be used to detect and identify pedestrians more accurately and efficiently, thereby improving the accuracy of the statistical results.

Among them, considering that when the camera collects the first image, due to the tilt angle of the camera, the number of pixels corresponding to a pedestrian close to the camera position is greater than the number of pixels corresponding to the position far from the camera position, that is, when the number of pedestrians in the first pixel area and the second pixel area is the same, the number of first target pixels is greater than the number of second target pixels. Therefore, in some embodiments, when determining the pedestrian number interval of the second pixel area based on the number of first target pixels, the number of second target pixels, and the number of pedestrians in the first pixel area, the ratio between the number of first target pixels and the number of second target pixels can be corrected. For example, compensation is performed based on the tilt angle parameter comparison value of the camera that captures the first image to improve the accuracy of the pedestrian number prediction result.

In some optional embodiments, determining the number of pedestrians in the first image based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area corresponding to the pedestrian number interval includes:

Obtaining a first predicted number of pedestrians in the first image according to the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area;

When the difference between the first predicted number and the second predicted number is less than or equal to a preset threshold, determining the first predicted number as the number of pedestrians in the first image;

The second predicted number is the number of pedestrians determined based on the historical number of pedestrians and external factors, and the external factors include weather factors and time factors when the acquisition device acquires the first image.

In this embodiment, as shown in FIG2 , the first image acquired by the acquisition device is first input into the object detection algorithm model based on Yolo to perform quantitative detection of pedestrians in the first image and obtain the number of pedestrians in the first pixel area. At the same time, the second pixel area in the first image that cannot be recognized by the object detection algorithm model is determined, and the second pixel area of the first image is further qualitatively detected by the image classification algorithm model based on ResNet to determine the number interval of pedestrians in the second pixel area. The pedestrian counting problem is converted into a classification problem, which reduces the difficulty of determining the number of pedestrians in the first image. The number of pedestrians in the first image is then determined based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area corresponding to the pedestrian number interval. In this way, the image classification algorithm and the object detection algorithm are combined to reduce the statistical error caused by omitting the second pixel area that cannot be recognized by the object detection algorithm, thereby improving the accuracy of the detection results.

Furthermore, the first predicted number of pedestrians in the first image is obtained based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area. Considering that there is a certain correlation between the number of pedestrians and external factors such as weather and time, for example, there will be more people during the morning and evening rush hours of the subway, there will be more people before lunch and dinner in the supermarket, and there will be more people on weekends than on weekdays. Therefore, the first predicted number of pedestrians in the first image can be verified based on the trained regression model to improve the accuracy of the detection result. Among them, the regression model is a verification model trained based on the historical number of pedestrians and sample data corresponding to external factors.

In one example, the weather factors and time factors when the acquisition device acquires the first image, as well as the historical number of pedestrians, can be input into the regression model to obtain the second predicted number; when the difference between the first predicted number and the second predicted number is less than or equal to the preset threshold, the first predicted number is determined as the number of pedestrians in the first image. By combining quantitative and qualitative methods to determine the number of pedestrians, and taking into account the influence of historical pedestrian data as well as weather factors and time factors, pedestrians can be detected and identified more accurately and efficiently, improving the accuracy of statistical results. This provides better technical support for public safety, intelligent transportation and other fields.

In another example, the weather factors and time factors when the acquisition device acquires the first image, as well as the first predicted number, can also be input into the regression model obtained after training to obtain a third predicted number. The third predicted number is determined as the number of pedestrians in the first image. By combining quantitative and qualitative methods to determine the number of pedestrians, and taking into account the influence of historical pedestrian data as well as weather factors and time factors, pedestrians can be detected and identified more accurately and efficiently, thereby improving the accuracy of statistical results. This provides better technical support for public safety, intelligent transportation and other fields.

Among them, if there is no second pixel area in the first image that cannot be recognized by the object detection algorithm model, then directly predicting the number of pedestrians in the first image and inputting it into the regression algorithm model can also achieve the same technical effect, which will not be repeated here.

Referring to FIG. 3 , FIG. 3 is a schematic diagram of the structure of a device for determining the number of pedestrians provided by an embodiment of the present invention. As shown in FIG. 3 , the device for determining the number of pedestrians 300 includes:

A first determination module 301 is used to determine the number of pedestrians in a first pixel area in a first image acquired by a collection device based on the first image;

A second determination module 302, configured to determine a pedestrian quantity interval of a second pixel area of the first image based on the second pixel area of the second pixel area;

A third determination module 303, configured to determine the number of pedestrians in the first image based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area corresponding to the pedestrian number interval;

The distance between the first position area mapped by the first pixel area and the acquisition device is smaller than the distance between the second position area mapped by the second pixel area and the acquisition device.

Optionally, the second determining module 302 includes:

The first determination submodule is used to determine the pedestrian number interval in the second pixel area according to the number of first target pixels, the number of second target pixels, and the number of pedestrians in the first pixel area, wherein the first target pixel is a pixel having a target feature in the first pixel area, the second target pixel is a pixel having the target feature in the second pixel area, and the target feature is a feature that characterizes pedestrians.

Optionally, the second determining module 302 includes:

The second determination submodule is used to detect the second pixel area of the first image according to the image classification algorithm, and determine the pedestrian quantity interval corresponding to the second pixel area in multiple preset pedestrian quantity intervals.

Optionally, the first determining module 301 includes:

A detection submodule, configured to detect the first image according to a Yolo object detection algorithm to obtain the number of pedestrians in the first pixel area;

The third determining submodule is used to determine a pixel region other than the first pixel region in the first image as the second pixel region.

Optionally, the first pixel area is a pixel area marked by the Yolo object detection algorithm in the first image.

Optionally, the third determining module 303 includes:

a fourth determination submodule, configured to obtain a first predicted number of pedestrians in the first image according to the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area;

a fifth determination submodule, configured to determine the first predicted number as the number of pedestrians in the first image when a difference between the first predicted number and the second predicted number is less than or equal to a preset threshold;

The second predicted number is the number of pedestrians determined based on the historical number of pedestrians and external factors, and the external factors include weather factors and time factors when the acquisition device acquires the first image.

The pedestrian number determination device 300 can implement each process of each embodiment of the above-mentioned pedestrian number determination method, the technical features correspond one to one, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

An embodiment of the present invention also provides an electronic device, including: a processor, a memory, and a program stored in the memory and executable on the processor. When the program is executed by the processor, the various processes of the above-mentioned fault prediction method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

Specifically, referring to FIG. 4 , an embodiment of the present invention further provides an electronic device, including a bus 401 , a transceiver 402 , an antenna 403 , a bus interface 404 , a processor 405 , and a memory 406 .

Wherein, the transceiver 402 is used to obtain the first image captured by the acquisition device;

Processor 405, configured to determine the number of pedestrians in a first pixel area in the first image based on the first image acquired by the acquisition device;

The processor 405 is further configured to determine a pedestrian quantity interval of the second pixel area based on the second pixel area of the first image;

The processor 405 is further configured to determine the number of pedestrians in the first image based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area corresponding to the pedestrian number interval;

The distance between the first position area mapped by the first pixel area and the acquisition device is smaller than the distance between the second position area mapped by the second pixel area and the acquisition device.

Optionally, the determining, based on the second pixel area of the first image, a pedestrian quantity interval of the second pixel area includes:

According to the number of first target pixels, the number of second target pixels, and the number of pedestrians in the first pixel area, the pedestrian number interval of the second pixel area is determined, the first target pixel is a pixel having a target feature in the first pixel area, the second target pixel is a pixel having the target feature in the second pixel area, and the target feature is a feature that characterizes pedestrians.

Optionally, the determining, based on the second pixel area of the first image, a pedestrian quantity interval of the second pixel area includes:

The second pixel area of the first image is detected according to an image classification algorithm, and a pedestrian quantity interval corresponding to the second pixel area is determined from a plurality of preset pedestrian quantity intervals.

Optionally, determining the number of pedestrians in a first pixel area in the first image based on the first image acquired by the acquisition device includes:

Detect the first image according to the Yolo object detection algorithm to obtain the number of pedestrians in the first pixel area;

A pixel region other than the first pixel region in the first image is determined as the second pixel region.

Optionally, the first pixel area is a pixel area marked by the Yolo object detection algorithm in the first image.

Optionally, the determining the number of pedestrians in the first image based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area corresponding to the pedestrian number interval includes:

Obtaining a first predicted number of pedestrians in the first image according to the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area;

When the difference between the first predicted number and the second predicted number is less than or equal to a preset threshold, determining the first predicted number as the number of pedestrians in the first image;

The second predicted number is the number of pedestrians determined based on the historical number of pedestrians and external factors, and the external factors include weather factors and time factors when the acquisition device acquires the first image.

In FIG. 4 , a bus architecture (represented by bus 401) is shown, and bus 401 may include any number of interconnected buses and bridges, and bus 401 links various circuits including one or more processors represented by processor 405 and memory represented by memory 406. Bus 401 may also link various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. Bus interface 404 provides an interface between bus 401 and transceiver 402. Transceiver 402 may be one element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices on a transmission medium. Data processed by processor 405 is transmitted on a wireless medium via antenna 403, and further, antenna 403 also receives data and transmits the data to processor 405.

Processor 405 is responsible for managing bus 401 and general processing, and may also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory 406 may be used to store data used by processor 405 when performing operations.

Optionally, the processor 405 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or a complex programmable logic device (CPLD).

The embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, each process of the above-mentioned pedestrian number determination method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. The computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

The embodiment of the present application also provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, the various processes of the above-mentioned pedestrian number determination method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, they are not repeated here.

It should be noted that, in this article, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the method and device in the embodiment of the present invention is not limited to performing functions in the order discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.

Claims (10)

1. A method for determining the number of pedestrians, characterized in that the method comprises: Determine, based on a first image acquired by an acquisition device, the number of pedestrians in a first pixel area in the first image; Determining a pedestrian quantity interval of a second pixel area based on a second pixel area of the first image; Determining the number of pedestrians in the first image based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area corresponding to the pedestrian number interval; The distance between the first position area mapped by the first pixel area and the acquisition device is smaller than the distance between the second position area mapped by the second pixel area and the acquisition device. 2. The method according to claim 1, characterized in that the determining the pedestrian quantity interval of the second pixel area based on the second pixel area of the first image comprises: According to the number of first target pixels, the number of second target pixels, and the number of pedestrians in the first pixel area, the pedestrian number interval of the second pixel area is determined, the first target pixel is a pixel having a target feature in the first pixel area, the second target pixel is a pixel having the target feature in the second pixel area, and the target feature is a feature that characterizes pedestrians. 3. The method according to claim 1, wherein determining the pedestrian quantity interval of the second pixel area based on the second pixel area of the first image comprises: The second pixel area of the first image is detected according to an image classification algorithm, and a pedestrian quantity interval corresponding to the second pixel area is determined from a plurality of preset pedestrian quantity intervals. 4. The method according to claim 1, characterized in that the determining the number of pedestrians in the first pixel area in the first image based on the first image acquired by the acquisition device comprises: Detect the first image according to the Yolo object detection algorithm to obtain the number of pedestrians in the first pixel area; A pixel region other than the first pixel region in the first image is determined as the second pixel region. 5. The method according to claim 4 is characterized in that the first pixel area is a pixel area marked by the Yolo object detection algorithm in the first image. 6. The method according to claim 1, characterized in that the determining the number of pedestrians in the first image based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area corresponding to the pedestrian number interval comprises: Obtaining a first predicted number of pedestrians in the first image according to the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area; When the difference between the first predicted number and the second predicted number is less than or equal to a preset threshold, determining the first predicted number as the number of pedestrians in the first image; The second predicted number is the number of pedestrians determined based on the historical number of pedestrians and external factors, and the external factors include weather factors and time factors when the acquisition device acquires the first image. 7. A device for determining the number of pedestrians, characterized in that the device comprises: A first determination module, configured to determine the number of pedestrians in a first pixel area in a first image acquired by an acquisition device based on the first image; A second determination module, configured to determine a pedestrian quantity interval of a second pixel area based on a second pixel area of the first image; a third determination module, configured to determine the number of pedestrians in the first image based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area corresponding to the pedestrian number interval; The distance between the first position area mapped by the first pixel area and the acquisition device is smaller than the distance between the second position area mapped by the second pixel area and the acquisition device. 8. An electronic device, comprising a transceiver and a processor, The processor is used to determine the number of pedestrians in a first pixel area in the first image based on the first image acquired by the acquisition device; The processor is further configured to determine a pedestrian quantity interval of the second pixel area based on the second pixel area of the first image; The processor is further configured to determine the number of pedestrians in the first image based on the number of pedestrians in the first pixel area and the number of pedestrians in the second pixel area corresponding to the pedestrian number interval; The distance between the first position area mapped by the first pixel area and the acquisition device is smaller than the distance between the second position area mapped by the second pixel area and the acquisition device. 9. A computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method for determining the number of pedestrians as described in any one of claims 1 to 6 are implemented. 10. A computer program product, characterized in that it comprises computer instructions, which, when executed by a processor, implement the steps of the method for determining the number of pedestrians according to any one of claims 1 to 6.