CN113297888B

CN113297888B - Image content detection result checking method and device

Info

Publication number: CN113297888B
Application number: CN202010986033.3A
Authority: CN
Inventors: 李松; 张文杰
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2020-09-18
Filing date: 2020-09-18
Publication date: 2024-06-07
Anticipated expiration: 2040-09-18
Also published as: CN113297888A

Abstract

The specification discloses a method and a device for checking an image content detection result. The method is realized based on the image content detection result set and comprises the following steps: and monitoring the repeated count value of the added result in the set, and if the repeated count value of any added result is greater than a preset threshold value, determining that the added result is an error detection result. When the set is updated, traversing the added results in the current set to be matched with the to-be-added results, and judging whether the added results successfully matched with the to-be-added results exist in the current set; if not, adding the result to be added into the current set, and initializing the repeated count value of the result; if so, performing increment processing on the repeated count values of the added results which are successfully matched; the repetition count is at least for: and when the repeated count value of any added result is larger than a preset threshold value, determining that the added result is an error detection result.

Description

Image content detection result checking method and device

Technical Field

The embodiment of the specification relates to the field of image processing, in particular to a method and a device for checking an image content detection result.

Background

Image content detection is a technique of detecting whether or not specified content exists in an image, and marking the specified content in the image. The image content detection technology is widely applied to monitoring scenes at present, the monitoring equipment can continuously shoot images of a designated monitoring area, and designated contents such as people, vehicles and the like existing in the monitoring area can be identified through the content detection technology so as to meet further application requirements.

Since the content detection is performed based on the image characteristics of the detection target, when there is a portion having some image characteristics but not specified content in the image, an unexpected detection result occurs. For example, when a person is detected in a monitoring image of a certain area, if a person poster happens to appear in the captured image, the position of the poster in the image is erroneously marked. Such erroneous detection results necessarily affect the accuracy of the subsequent application.

Disclosure of Invention

In order to check out an error detection result and improve the accuracy of image content detection, the specification provides a method and a device for checking the image content detection result. The specific technical scheme is as follows.

An image content detection result set updating method includes the steps that a content detection result set for checking is established for a monitoring image of any area, and each added result in the set has a repeated count value; the method comprises the following steps:

After any monitoring image of the area is obtained, content detection is carried out on the image, and the obtained content detection result is determined as a result to be added;

Traversing the added results in the current content detection result set to be matched with the to-be-added results, and judging whether the added results successfully matched with the to-be-added results exist in the current set; if not, adding the result to be added to the current set, and initializing a repetition count value of the result; if so, performing increment processing on repeated count values of all the successfully matched added results according to a preset increment rule;

The repetition count value is at least for: and when the repeated count value of any added result is larger than a preset threshold value, determining that the added result is an error detection result.

An image content detection result checking method based on the set in the image content detection result set updating method comprises the following steps:

And monitoring the repeated count value of the added result in the set, and if the repeated count value of any added result is greater than a preset threshold value, determining that the added result is an error detection result.

An image content detection result set updating device creates a content detection result set for checking for a monitoring image of any area, wherein each added result in the set has a repeated count value; the device comprises:

and a detection unit: after any monitoring image of the area is obtained, content detection is carried out on the image, and the obtained content detection result is determined as a result to be added;

Matching unit: traversing the added results in the current content detection result set to be matched with the to-be-added results, and judging whether the added results successfully matched with the to-be-added results exist in the current set; if not, adding the result to be added to the current set, and initializing a repetition count value of the result; if so, performing increment processing on repeated count values of all the successfully matched added results according to a preset increment rule;

An image content detection result checking device based on the set in the image content detection result set updating device, comprising:

monitoring unit: monitoring the repeated count value of the added result in the set;

A judging unit: if the repeated count value of any added result is larger than the preset threshold value, determining that the added result is an error detection result.

According to the technical scheme, the repeated appointed content in the multiple images is determined by constructing and updating the image content detection result set and utilizing the attribute value of the repeated count value. When the repetition count value is greater than the threshold value, the dynamic characteristics of the specified content can be based, that is, the correct specified content does not repeatedly appear in a plurality of monitoring images with the number greater than the threshold value, so that the corresponding specified content is determined to be an error detection result, and the accuracy of image content detection is improved.

Drawings

In order to more clearly illustrate the embodiments of the present description or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the embodiments of the present description, and other drawings may be obtained according to these drawings for a person having ordinary skill in the art.

Fig. 1 is a schematic diagram of a pedestrian detection method according to an embodiment of the present disclosure;

Fig. 2 is a schematic diagram of pedestrian detection in a monitoring scene according to an embodiment of the present disclosure;

Fig. 3 is a flowchart of an updating method for an image content detection result set according to an embodiment of the present disclosure;

fig. 4 is a flowchart of a method for checking an image content detection result according to an embodiment of the present disclosure;

Fig. 5 is a schematic diagram of an application example of an image content detection result checking method according to an embodiment of the present disclosure;

fig. 6 is a schematic structural diagram of an apparatus for updating an image content detection result set according to an embodiment of the present disclosure;

fig. 7 is a schematic structural diagram of an image content detection result checking device provided in the embodiment of the present specification;

Fig. 8 is a schematic structural view of an apparatus for configuring the method of the embodiment of the present specification.

Detailed Description

In order for those skilled in the art to better understand the technical solutions in the embodiments of the present specification, the technical solutions in the embodiments of the present specification will be described in detail below with reference to the drawings in the embodiments of the present specification, and it is apparent that the described embodiments are only some embodiments of the present specification, not all embodiments. All other embodiments obtained by a person skilled in the art based on the embodiments in the present specification shall fall within the scope of protection.

In the present specification, image content detection may be a technique of detecting whether or not specified content exists in an image and marking the specified content in the image.

The image content detection technique, when implemented, may be to first extract content features in the image, and if the extracted content features include features of the specified content, it may be determined that the specified content exists in the image, and the position of the specified content in the image may be further marked.

The present disclosure is not limited to a specific technique for extracting image content features, and an alternative embodiment may be implemented using a deep neural network model, where implicit image content features are extracted for an image using a trained deep neural network model.

There are various application scenarios of the image content detection technology, and specific objects to be detected may also be different, for example, when the specified content is a pedestrian, the specific image content detection technology may be regarded as a pedestrian detection technology; when the specified content is a vehicle, the specific image content detection technique may be regarded as a vehicle detection technique.

Taking a pedestrian detection scene as an example, as shown in fig. 1, a schematic diagram of pedestrian detection is provided in the present specification.

The pedestrian detection technology is used for detecting one image, 3 pedestrians in the image can be detected, positions of the detected pedestrians in the image are obtained, and the positions of the 3 pedestrians in the image are marked by using rectangular frames.

It should be noted that, for any image, the detection result may include one or more specified contents, and may also include 0 specified contents. Through the image content detection technology, under the condition that N appointed contents exist in an image, the positions of the N appointed contents in the image can be obtained, wherein N is more than or equal to 0.

The image content detection technology is widely applied to monitoring scenes at present, the monitoring equipment can continuously shoot images of a designated monitoring area, and designated contents such as people, vehicles and the like existing in the monitoring area can be identified through the content detection technology so as to meet further application requirements.

When the monitoring device continuously shoots images, the specification does not limit whether the interval time period between any two adjacent image shooting is fixed or not, and does not limit the frequency of image shooting.

And the designated monitoring area continuously photographed by the same monitoring device may be the same geographical area.

Therefore, the same monitoring device continuously shoots and acquires a plurality of images aiming at the same geographic area.

When image content detection is performed on a single image acquired by the monitoring device, the position of the acquired specified content in the image can be characterized by the position of the specified content in the geographic area shot by the monitoring device at the moment the monitoring device shoots the image.

When image content detection is performed on a plurality of images acquired by the monitoring device, the positions of the acquired specified contents in the images can be represented by the positions of the specified contents in the geographic areas shot by the monitoring device at a plurality of moments when the images are shot by the monitoring device.

In other words, when the image content detection technology is applied to a monitoring scene, the image content detection technology is used for detecting a plurality of images continuously acquired by the same monitoring device, so that the positions of the appointed content in the geographic area shot by the monitoring device at a plurality of moments can be acquired, and further application requirements can be met through the positions of the appointed content at the plurality of moments.

Fig. 2 is a schematic diagram of pedestrian detection in a monitoring scene provided in the present specification.

The appointed monitoring areas shot by the monitoring equipment in the figure are 3 shelves in a store, namely a commodity shelf, a food shelf and a refrigerator shelf.

The monitoring device continuously acquires 3 images, which are respectively shot at 6:00 am, 12:00 pm and 3:00 pm. The pedestrian detection is carried out on 3 images continuously acquired by the monitoring equipment, so that the positions of the pedestrians in the 3 images can be obtained respectively, and the positions of the pedestrians in the appointed monitoring area can be obtained at 3 moments of 6:00 am, 12:00 pm and 3:00 pm respectively.

In the monitoring scene, the positions of the specified contents at a plurality of moments, which are acquired by the image content detection technology, can be further used for different application requirements.

The application demands are, for example, calculation of motion trajectories of the specified contents, statistics of the number of vehicles at different times, analysis of people flow distribution at different times, and the like.

For ease of understanding, a practical example of application is given below for illustrative purposes.

In a new retail scene, people and goods relationship is often required to be dataized so as to optimize retail goods distribution, goods-in quantity and the like by utilizing leading edge technologies such as big data analysis, data mining and the like, and improve resource utilization rate, transaction occurrence rate and the like.

The person-goods relationship data includes, for example, the residence time of pedestrians before different goods, the try-on time, the types and the number of goods purchased by pedestrians of different ages, the probability of the pedestrians who have purchased one type of goods purchasing other types of goods near the goods, and the like.

Using these data-based man-machine relationships, a number of conclusions can be drawn, for example, a person purchasing beer will often purchase snacks and other consumables, and the analyzed conclusions can be used to optimize the distribution of the goods, and so forth.

The commodity taking behavior identification is an important part of the human-to-commodity relationship data, and is used for identifying commodity taking and fitting behaviors of pedestrians. One of the more important steps is pedestrian detection.

The intelligent terminal arranged in the online store can shoot a shelf picture through the monitoring camera, the pedestrian detection is completed through image recognition, the existence of the pedestrian in the image is detected, and the position of the pedestrian in the image is determined, so that the action detection is further executed.

If the detected motion is a motion of picking up a commodity, it can be considered that a commodity picking up behavior has occurred to a pedestrian.

However, since the content detection may be implemented based on the image characteristics of the specified content, when there is some image characteristic having the specified content in the image, but the content other than the specified content, a detection result that does not conform to expectations, that is, a false detection result, may occur.

For example, when detecting pedestrians in a monitored image of a certain area, if a person poster happens to appear in the captured image, since the person on the poster also has the content characteristics of the pedestrians, the person on the poster in the image is erroneously detected as a pedestrian, and the position of the person on the poster is marked. Such erroneous detection results necessarily affect the accuracy of the subsequent application.

For another example, in the case of detecting a vehicle in a monitoring image of a certain area, if a car poster happens to appear in the captured image, the position of the poster in the image is erroneously marked. Such erroneous detection results necessarily affect the accuracy of the subsequent application.

Also for example, in the new retail scenario described above, there often occur in the store objects in which a person poster, a humanoid display board, a mannequin, a dynamic person in the screen, and the like are easily erroneously detected as pedestrians. When using the pedestrian detection technique, if the accuracy of pedestrian detection is low, the subsequent analysis conclusion is not accurate.

In order to improve the accuracy of the image content detection technology, the specification provides an image content detection result checking method.

By analyzing the error detection results detected by the image content detection technology under different application scenes, it is found that most of the error detection results have the content characteristics of the specified content, but the position can be kept unchanged or the characteristics can be kept unchanged. Such as posters, models, etc.

From a real scene, the poster posted in the store is often fixed in position, and in a plurality of images continuously acquired by the monitoring device, the position of the poster in the image should also be fixed. Similarly, human-shaped display boards, manikins, dynamic characters in screens, and the like in shops are easily erroneously detected as objects of pedestrians, and are usually placed at fixed positions, and the positions in images should be fixed.

Even if a poster posted within a store changes the location of the post, the character content on the poster does not change, and thus the corresponding content characteristics may be unchanged.

While the specified content may be generally dynamic, with dynamic characteristics, it is difficult to keep the location or characteristics unchanged for a long period of time. The dynamic characteristic can be explained in particular in that the specified content can be dynamic in a normal state, can change positions, does not remain stationary for a long time, and does not repeatedly appear in the same area.

For example, the posture and position of the pedestrian may be changed in general, and may not remain stationary for a long period of time, and thus the pedestrian has a dynamic characteristic.

Through analysis of the error detection results and dynamic characteristics of the specified content, determining the commonality of the error detection results may include: repeated occurrences are within the same area, i.e. repeated occurrences are in more monitored images. The judgment can be specifically performed through the position or the content characteristics.

Therefore, the method for checking the image content detection result provided by the specification is mainly used for checking whether the detection result is the designated content or not by checking whether the detection result is always kept at a fixed position or always has the same content characteristics in a plurality of monitoring images shot by monitoring equipment aiming at the designated content with dynamic characteristics, so that the error detection result is eliminated, and the accuracy of image content detection is improved.

It should be noted that, in the method for checking an image content detection result provided in the present specification, implementation of the technical effect of improving accuracy of an image content detection technique may be based on that specified content has dynamic characteristics.

For further understanding, the image content detection result checking method provided in the present specification is explained in detail below with reference to the accompanying drawings.

Since it is necessary to determine whether there is a detection result that remains stationary, such as a position-invariant or a content feature-invariant, in the plurality of images, and each detection is an image content detection for a single image captured by the monitoring device, it is necessary to accumulate historical image content detection results in the method, and check whether there is an erroneous detection result based on the accumulated plurality of images.

Therefore, in the present specification, for the monitoring image of any area, a content detection result set for checking the error detection result can be created for storing the image content detection result of the history. The content detection result set may also be updated continuously according to the newly detected detection result. Based on the updated content detection result set, the detection result can be checked.

The present disclosure provides an image content detection result set updating method, configured to continuously update an image content detection result set according to a new detection result, so as to implement an image content detection result checking method according to a currently updated image content detection result set, thereby enabling a check of a detection result to be performed in real time.

Fig. 3 is a schematic flow chart of an updating method for image content detection result set provided in the present specification. For a monitored image of any region, a set of content detection results for verification may be created, each added result in the set may have a repetition count value. Of course, the content detection result set just created may be an empty set.

The method may comprise the following steps.

S101: after any monitoring image of the area is obtained, content detection is carried out on the image, and the obtained content detection result is determined as a result to be added.

S102: and traversing the added results in the current content detection result set to be matched with the results to be added.

S103: judging whether an added result successfully matched with the result to be added exists in the current set.

If not, S104 is performed. If so, S105 is performed.

S104: and adding the result to be added to the current set, and initializing the repeated count value of the result.

S105: and performing increment processing on the repeated count values of all the successfully matched added results according to a preset increment rule.

Wherein the repetition count value may be used at least for: and when the repeated count value of any added result is larger than a preset threshold value, determining that the added result is an error detection result.

The following is a description of the overall embodiment.

It should be noted that, since the monitoring image for any one area can create the content detection result set, the monitoring images for other areas can also be created and updated by the above-described method.

In addition, since the created and updated content detection result set is used for verification, the content detection result is not simply stored, and thus, for repeated detection results, the repetition count value included in the added detection result in the set can be directly increased, and is not required to be added to the set.

The repetition count value may be one attribute value of the added result in the set, or may be one attribute value corresponding to the added result outside the set, which is not limited in this embodiment. The repetition count value can be specifically considered to represent the number of monitoring images in which the detection result repeatedly appears.

For the sake of facilitating understanding of the present embodiment, the repetition count value may be further understood as a mechanism for checking the detection result that remains stationary in the set, and when any added result in the set matches successfully with the to-be-added result, it may be explained that the added result is identical to the to-be-added result in terms of the position in the monitored image or the content feature in the image, and therefore, the to-be-added result may not be added to the set and the repetition count value of the added result may be increased.

Of course, the correct detection result may also repeatedly appear in a plurality of images, and remain stationary, for example, the pedestrian pauses the steps, or the traveling vehicle temporarily stops at the side, but the pedestrian does not pause the steps for a whole day, and the vehicle traveling on the expressway does not stop for a whole day. Therefore, in the mechanism of repeating the count value, a preset threshold value needs to be added to distinguish between a correct detection result and an incorrect detection result.

When the repeated detection results are more and more than the preset threshold value, the positions of the detection results in a plurality of images with more than the threshold value can be considered to be unchanged, or the characteristics in a plurality of images with more than the threshold value can be considered to be unchanged, namely the detection results do not accord with the dynamic characteristics of the designated content and do not belong to the designated content, and the detection results can be determined to be error detection results.

That is, as mentioned above, the repetition count value may be used at least for: and when the repeated count value of any added result is larger than a preset threshold value, determining that the added result is an error detection result.

Obviously, by repeating the counting value mechanism, the detection result which is kept static can be checked in the set, and then the error detection result is determined.

In the repetition count value, the specification is not limited, and specifically, the error detection result is determined by accumulating to a preset threshold value or by decrementing to the preset threshold value. So long as the detection results that remain stationary can be checked in the collection.

It should be noted that, in this specification, more preset thresholds will occur, and the preset thresholds in different mechanisms may be different from each other.

After explaining the overall embodiment, explanation is made below for each step.

For S101, it should be explained that, firstly, any monitoring image of the area may be acquired according to the shooting sequence of the monitoring device, that is, the time sequence of shooting by the monitoring device; or may be acquired out of order, for example, by acquiring the currently captured image first and then acquiring the previously captured image. The present specification is not limited thereto. This is because the update of the image content detection result set can be realized and the check of the detection result can be performed no matter in which order the monitoring images of the region are acquired.

In the case of content detection for an image, the present embodiment is not limited to the algorithm used to specifically implement the content detection technology, and may be a deep neural network.

The obtained content detection result may be a characteristic value specifying the position of the content in the image, or a characteristic value specifying the position of the content in the image, and the embodiment is not limited thereto.

As a specific example, the content detection result may be in the form of specifying coordinate positions of four vertices of a rectangular frame in an image after the content is marked by the rectangular frame. These four coordinates may be used to characterize the location of the specified content in the image.

It can be understood that when the detection result represents the position of the specified content in the image, the storage space occupied by the representation value is smaller, so that the storage space is saved, the subsequent matching is convenient, and the efficiency of traversing the matching in S102 can be improved.

When the detection result represents the characteristic of the appointed content in the image, the characteristic value per se contains more information, and the detection result can be directly and accurately checked.

Of course, as an alternative embodiment, the characterization value of the position of the specified content in the image and the characterization value of the specified content in the image may be combined to be used as the content detection result, and the advantages of the two characterization values may be combined. For example, the position characterization value is matched first, so that the matching speed is increased, and the feature characterization value can be matched further under the condition that the matching is successful, so that the matching accuracy is improved.

Further, since a plurality of specified contents can be detected by the image content detection technique in a single image. Therefore, in the case where a plurality of specified contents are detected, a plurality of detection results can be obtained, and the steps of S101 to S105 described above are performed for each detection result.

It is to be noted that although the content detection result is determined as a result to be added, it is not necessarily added to the collection. In S105, in the case where there is an added result in the set that successfully matches the result to be added, the result to be added may not be added to the set.

For S102, after the to-be-added result is matched with any added result, the to-be-added result can be matched with other added results no matter what the matching result is, and the traversing step is not stopped by the matching result. After matching the to-be-added result with all the added results in the current set, the following steps may be performed.

It should be noted that when matching is performed, matching needs to be performed with respect to the current set in order to check the error detection result in real time.

And in particular, the matching can be that the result to be added is compared with any added result, and when the compared difference characterization value is smaller than a preset threshold value, the matching is successful; and when the matching time is greater than a preset threshold value, the matching fails.

Of course, the preset threshold value here and the preset threshold value in the repetition count value mechanism may be different. The preset thresholds in different meanings in the present specification may not be the same specific values.

The difference representing value can specifically represent the distance between the to-be-added result and the position of any added result in the image, and can also represent the difference between the to-be-added result and the feature of any added result in the image.

An example of the gap characterization value is given below.

The gap characterization value may characterize a distance between the to-be-added result and a position of any added result in the image, and the detection result may be in the form of four vertex coordinates of a rectangular frame characterizing a position of the specified content in the image.

The calculation of the difference representation value may specifically be that the ratio between the area of the overlapping part of the rectangle 1 and the rectangle 2 and the sum of the areas of the rectangle 1 and the rectangle 2 is determined according to the rectangle 1 surrounded by the four vertex coordinates of the result to be added and the rectangle 2 surrounded by the four vertex coordinates of any added result.

Obviously, there may be multiple added results in the current set that match successfully with the result to be added. In S105, the increment processing may be performed for the repetition count value of the added result that all matches succeed.

For S104, after adding the result to be added to the set, the repetition count value of the result to be added may be initialized to a fixed value. The present embodiment is not limited to a specific initial value of the repetition count value.

For S105, the embodiment is also not limited to a specific preset increment rule and a preset threshold corresponding to the repetition count value.

Two examples of preset delta rules are given below for illustrative purposes.

1) The repetition count value may represent the number of repetitions, and the increment rule may be that each of the added results matches a result to be added successfully, the repetition count value being incremented by 1.

2) The repetition count value may represent a duration of rest and the increment rule may be dependent on the frequency with which the monitoring device captures images. For example, when the monitoring device captures 1 image every 2 seconds, the increment rule may be that each of the added results matches one of the results to be added successfully, and the repetition count value is incremented by 2.

According to the embodiment of the method, the corresponding image content detection result set can be updated aiming at one monitoring area, and the error detection result in the set is determined by using the repeated count value, so that the error detection result can be checked by using the current updated set, and the accuracy of image content detection is improved.

In other alternative method embodiments, the historical image content detection result may also be stored in other forms, such as a table. The table may include at least the detection result and the corresponding repetition count value. For further steps reference is made to the method embodiments described above.

Furthermore, in order to accelerate the execution efficiency of the traversal in S102 in the above method embodiment, save the storage resources of the set, delete the fewer detection results in the set, and reduce the number of detection results contained in the set.

Similar to the mechanism of repetition count values described above, in an alternative embodiment, by using the mechanism of matching failure count value, in the case that any added result in the set fails to match with the result to be added, the matching failure count value is increased until the value is greater than the preset threshold, and then the added result is considered to be unsuccessful in matching for a long time, that is, the detection result in the set is not updated for a long time, and the added result may be deleted from the current set.

Of course, similar to the repetition count value, the present embodiment does not limit the mechanism of matching the failure count value, specifically, whether to accumulate to the preset threshold or to decrement to the preset threshold. So long as it can indicate whether the added result has not been successfully matched for a long time.

The matching failure count value is explained in more detail below.

Based on the method embodiment shown in fig. 3, described above, each added result in the set may also have a match failure count value.

If any added result fails to match with the result to be added, the matching failure count value of the added result can be subjected to incremental processing according to a preset incremental rule. The increment rule here may be different from that of the repetition count value.

The match failure count value may be used at least for: and deleting any added result from the current set after the matching failure count value of the added result is larger than a preset threshold value.

Through the above steps, the matching failure count value can be used as a mechanism for deleting the detection results which are less used in the set.

Further, when the added result is successfully matched with the to-be-added result, the image content corresponding to the added result is shown to be in the monitoring image, and the to-be-added result is updated, so that the deleting speed can be reduced.

Therefore, in order to slow down the deletion, in the case where there is an added result that successfully matches the result to be added in the set, the match failure count value of the added result that is successfully matched in its entirety may be decremented according to a preset decrement rule.

The embodiment does not limit specific increment rules and decrement rules in the mechanism of matching failure count values, and examples of increment rules may refer to the explanation of the repetition count values described above. Two examples are given below for the decrementing rule for illustrative purposes.

1) The match failure count value may represent the number of match failures, and then the decrement rule may be that each of the added results and one of the to-be-added results match successfully, and the match failure count value is decremented by 1.

2) The match failure count value may represent a duration of the match failure and the decrement rule may be dependent on the frequency with which the monitoring device captures the image. For example, when the monitoring device captures 1 image every 2 seconds, the decrement rule may be that each of the added results and one of the to-be-added results match successfully, and the match failure count value is decremented by 2.

By the mechanism of matching the failure count value, the number of detection results in the set can be reduced to a certain extent, the execution efficiency of the detection results in the traversing set in the step S102 in the embodiment of the method is improved, and meanwhile, the storage resources of the set are saved.

It should be noted that, although some of the detection results in the set may be deleted by matching the failure count value, the detection result that was not determined to be the erroneous detection result before deletion may be used as the correct detection result in the subsequent application. Of course, the detection result determined as the error detection result before deletion may be used as the error detection result in a subsequent application.

And further considering the added result determined as the error detection result and the added result not determined as the error detection result, respectively, since the set is for checking the detection result, and more checking is performed with the added result determined as the error detection result, the added result determined as the error detection result can be stored in the set for a longer time for facilitating the subsequent checking.

In order to cause the error detection result to be stored in the set for a longer time, based on the mechanism of the above-described matching failure count value, two modified examples are provided below for illustrative purposes.

1) The match failure count value of the added result determined as the error detection result may be deleted from the current set after being greater than the first preset threshold. And the match failure count value of the added result that is not determined to be the error detection result may be deleted from the current set after being greater than the second preset threshold.

Wherein the first preset threshold may be greater than the second preset threshold in order to leave the added result determined as the false detection result in the set for a longer time.

2) The added result may reset the match failure count value to 0 when determined as an error detection result, in order to leave the added result determined as an error detection result in the set for a longer time.

Of course, the present specification is not limited to a specific manner of adjusting the matching failure count value mechanism of the error detection result, as long as the added result determined as the error detection result can be left in the set for a longer time, and is within the scope of the disclosure of the present specification.

In addition to improvements based on the match failure count value mechanism, in another alternative embodiment, in order to have the error detection result stored in the set for a longer time, two different deletion mechanisms may be used for the error detection result and the added result that is not determined to be the error detection result, respectively.

For the added result that is not determined as the error detection result, a mechanism of valid time may be used, and the valid time of the added result may be reduced according to a preset decrement rule when the matching fails until the added result is reduced to a preset threshold value, and the added result is deleted.

And for the added result determined as the error detection result, a mechanism of vanishing time may be used, and the vanishing time of the error detection result may be increased according to a preset increment rule when the matching fails until a preset threshold is increased, and the added result is deleted.

Similar to the mechanism of the matching failure count value, the present specification does not limit the specific mechanism to the valid time or the vanishing time, as long as the corresponding added result can be deleted after the matching failure is accumulated for a certain number of times, as long as the preset threshold is accumulated or decremented. The corresponding preset increment rule or decrement rule is not limited in this specification, and specific examples may refer to the above explanation of the repetition count value and the match failure count value.

It is to be noted that the valid time may not be effective while the added result, which is not determined as the error detection result, is determined as the error detection result, and the vanishing time may be further initialized for the subsequent deletion step.

The effective time is explained in detail below.

Based on the method embodiment shown in FIG. 3 above, each added result in the collection may also have a valid time.

If any added result fails to match with the result to be added, the effective time of the added result can be subjected to decrement processing according to a preset decrement rule.

Of course, the validity time may also decrease over time, regardless of whether the match failed.

The effective time can be at least used to: when the effective time of any added result is 0, the added result is deleted from the current set.

Furthermore, if there is an added result successfully matched with the result to be added, the effective time of all the successfully matched added results can be subjected to incremental processing according to a preset incremental rule, so that the deleting speed can be reduced.

The vanishing time is explained in detail below.

Based on the method embodiment shown in fig. 3 above, each added result in the set that is determined to be a false detection result may also have a vanishing time.

If any added result fails to match with the result to be added, incremental processing can be performed on the vanishing time of the added result according to a preset incremental rule.

The vanishing time can be used at least for: and deleting any added result from the current set after the vanishing time of the added result is greater than a preset threshold.

Further, if there is an added result that is successfully matched with the result to be added, the disappearance time of the added result that is successfully matched in its entirety and is determined as the error detection result may be subjected to the decrement processing according to a preset decrement rule.

It should be noted that, the matching failure count value, the valid time, and the vanishing time mentioned in the above embodiment may be attribute values included in the added result in the set, or may be attribute values corresponding to the added result outside the set.

When the form of storing the detection result is a table, the matching failure count value, the valid time, and the vanishing time may be column names in the table.

Based on the updated set in the above method embodiment, historical detection results may be stored for determining an error detection result according to the repetition count value. Furthermore, the number of the detection results in the set can be reduced according to the matching failure count value, or the effective time and the vanishing time, so that the execution efficiency of the detection results in the traversing set can be improved, and meanwhile, the storage resources of the set can be saved.

By using the content detection result set obtained by current updating in the embodiment of the method, the specification also provides a method for checking the image content detection result.

Fig. 4 is a schematic flow chart of a method for checking image content detection results provided in the present specification. Based on the content detection result set currently updated in any of the above method embodiments, corresponding to the mechanism of repeating the count value, the checking method may at least include the following steps.

S201: and monitoring the repeated count value of the added result in the set.

S202: if the repeated count value of any added result is larger than the preset threshold value, determining that the added result is an error detection result.

S203: for any to-be-added result, if the added result successfully matched in the set is an error detection result, the to-be-added result is determined to be the error detection result.

Wherein, in S201, monitoring of the repetition count value of the added result in the set is maintained. Upon monitoring that the repetition count value of any one of the added results is greater than the preset threshold, S202 is executed to determine that the added result is an error detection result.

While S203 does not need to rely on S201 or S202 to execute, it may be executed directly from the set, and thus S203 may be executed in parallel with S201.

Based on the embodiment of the method, the error detection result can be checked according to the content detection set, so that the error detection result can be determined conveniently, the accuracy of the image content detection technology is improved, and the accuracy of subsequent application is also improved.

In order to embody the method embodiment and improve the accuracy of subsequent application, the specification also provides two method embodiments in specific application scenes, namely a method embodiment in a passenger flow statistics scene and a method embodiment in a pedestrian attention behavior recognition scene.

The passenger flow statistics scene can be specifically that the number of pedestrians and pedestrians is counted by using monitoring equipment installed in places such as a mall, a supermarket or a store. The statistical result can be particularly used for analyzing specific businesses such as passenger flow distribution, hot goods or hot stores.

Obviously, based on traditional pedestrian detection, people in the poster, store models and the like are often identified as pedestrians during passenger flow statistics, so that the accuracy of passenger flow statistics is affected. Based on the embodiment of the method, normal pedestrians can be identified more accurately by screening out the error detection result, and the accuracy of passenger flow statistics is improved.

Therefore, the embodiment of the method in the passenger flow statistics scene may specifically include a detection method for pedestrian detection results and a method for passenger flow statistics by using an error detection result in the pedestrian detection results.

Based on the above method embodiment, the detection method for the pedestrian detection result may specifically be a pedestrian detection result set updating method, and for the monitored image of any area, a pedestrian detection result set for verification is created, where each added result in the set has a repetition count value. The method may comprise at least the following steps.

S301: after any monitoring image of the area is obtained, pedestrian detection is carried out on the image, and the obtained pedestrian detection result is determined as a result to be added.

S302: traversing the added results in the current pedestrian detection result set to be matched with the results to be added, and judging whether the added results successfully matched with the results to be added exist in the current set; if not, adding the result to be added into the current set, and initializing the repeated count value of the result; if so, performing increment processing on the repeated count values of all the successfully matched added results according to a preset increment rule.

The repetition count value may be used at least for: and when the repeated count value of any added result is larger than a preset threshold value, determining that the added result is an error detection result.

A passenger flow statistics method can include at least the following steps using the detected false detection result.

S401: updating the pedestrian detection result set and determining an error detection result.

The updating step may specifically be S301-S302. Multiple updates can be performed for the pedestrian detection result set, and then the error detection result is determined.

S402: after any monitoring image of the area is obtained, pedestrian detection is carried out on the image, and a pedestrian detection result is obtained.

S403: and determining the false detection result in the pedestrian detection results according to the false detection results in the set, and counting the number of the non-false detection results in the pedestrian detection results for passenger flow statistics.

In the embodiment of the method under the traffic statistics scene, the monitoring image specific to any area may be a monitoring image specific to an area such as a store, a mall, a traffic station, and the like.

After the false detection result in the pedestrian detection result is screened out, the counted pedestrians are basically normal pedestrians, so that the accuracy of the passenger flow counting method realized by utilizing the pedestrian detection is improved.

The pedestrian attention behavior recognition scene may specifically be that pedestrian attention behavior recognition is performed by using monitoring equipment installed in a place such as a mall, a supermarket, or a store.

The pedestrian attention behavior at least comprises commodity taking behavior, commodity trial behavior and try-on behavior, and further determines commodities in which the pedestrian is interested.

After the pedestrian attention behavior and the commodity of the pedestrian attention are identified, specific business analysis can be performed, for example, the commodity of the pedestrian attention is analyzed, or whether the commodity is actually purchased after the pedestrian attention is combined with the actual purchasing behavior is analyzed, so that a sales strategy or a product strategy is adjusted, and the like.

In a more specific scenario, the article of footwear may be identified for a pick-up and try-on activity.

The pedestrian attention behavior recognition may specifically be to further recognize whether the pedestrian generates the pedestrian attention behavior by the image recognition technology on the basis of detecting the pedestrian in the monitored image by using the conventional pedestrian detection technology.

However, the pedestrian detection often detects false detection results such as poster characters, store models and the like, and characters corresponding to the false detection results often have pedestrian attention behaviors, so that the accuracy of recognition results is low.

A pedestrian attention behavior recognition method may include at least the following steps using the detected false detection result.

S501: updating the pedestrian detection result set and determining an error detection result.

S502: after any monitoring image of the area is obtained, pedestrian detection is carried out on the image, and a pedestrian detection result is obtained.

S503: and determining the false detection result contained in the pedestrian detection result according to the false detection result in the set.

S504: for any one of the non-false detection results, the image recognition technology is utilized to recognize whether the behavior of the pedestrian corresponding to the non-false detection result belongs to any one of the pedestrian attention behaviors.

Obviously, after the false detection result in the pedestrian detection result is screened out, the identified pedestrians are basically normal pedestrians, so that the accuracy of the pedestrian attention behavior identification method based on pedestrian detection is improved.

In order to facilitate further understanding, the present specification also provides an application example of the image content detection result checking method.

The specific application scene can be that under the new retail scene, pedestrian distribution conditions of pedestrians in front of different types of commodities at different moments are determined through pedestrian detection, and then the most popular commodity types are determined according to the distribution conditions.

The appointed monitoring area shot by the monitoring equipment is 3 shelves in the store, namely a commodity shelf, a food shelf and a refrigerator shelf. The monitoring device takes an image every 2 seconds for pedestrian detection.

And constructing an image content detection result set in advance according to the designated monitoring area, detecting pedestrians according to the latest shot image of the monitoring equipment in time, and updating the set according to the detection result.

In order to facilitate the representation of the set, each detection result in the set is shown in the form of a table, and the table contains a repetition count value and a matching failure count value in addition to the detection result. The detection result is represented by a representation value of a position in the image, and of course, for convenience of representation, a single position coordinate in the image may be specifically used for representation. The repeated count value and the failed match count value can both adopt a mechanism of accumulating to a preset threshold value, the preset threshold value of the repeated count value is 10, and the preset threshold value of the failed match count value is 20.

A specific set of content detection results may be as shown in table 1 below.

Sequence number	Detection result	Repetition count value	Count value of failed match
				1	(5,5)	10	0
2	(1,4)	0	3
				3	(0,0)	0	5
4	(1,3)	0	20

TABLE 1 content detection result set

Fig. 5 is a schematic diagram of an application example of the image content detection result checking method provided in the present specification.

Wherein, pedestrian detection is performed based on the image in fig. 5, and 2 detection results are obtained, respectively, (5, 5) and (1, 2).

Aiming at the detection results (5, 5), the detection results 1-4 in the table 1 are traversed for matching, and the matching with the detection result 1 is obviously successful, and the matching with the detection results 2-4 is failed. Therefore, the repetition count value of the detection result 1 is incremented by 1, and the matching failure count values of the detection results 2 to 4 are all incremented by 1.

And summarizing the updated set, and determining that the repeated count value 11 of the detection result 1 is larger than the preset threshold value 10, wherein the detection result 1 is determined to be an error detection result. If the match failure count value 21 of the detection result 4 is greater than the preset threshold value 20, the detection result 4 is deleted from the set.

For the detection results (1, 2), the detection results 1-3 in the table 1 are traversed to match, and the matching with the detection results 1-3 is obviously failed. Therefore, the matching failure count values of the detection results 1 to 3 are all added by 1, and the detection results (1, 2) are added to the set as the detection result 5.

The updated content detection result set according to the above 2 detection results may be as shown in table 2 below.

Sequence number	Detection result	Repetition count value	Count value of failed match
				1	(5,5)	11	1
2	(1,4)	0	5
				3	(0,0)	0	7
5	(1,2)	0	0

TABLE 2 updated current content detection result set

Wherein, since the detection result 1 is determined as an erroneous detection result, the detection results 2 to 5 can be used as a correct detection result for subsequent data analysis.

From the perspective of a person, the detection result 1 is obviously a person poster, after the error detection result of the detection result 1 is eliminated, the rest correct detection result can help to improve the accuracy of data analysis, and the more accurate pedestrian distribution condition is obtained, so that the most popular commodity type is accurately determined.

In addition to the above-described method embodiments, the present specification also provides two apparatus embodiments, that is, an image content detection result set updating apparatus and an image content detection result checking apparatus, respectively.

As shown in fig. 6, a schematic structural diagram of an apparatus for updating an image content detection result set is provided in the present specification. For the monitoring image of any area, a content detection result set for verification is created, and each added result in the set has a repetition count value.

The image content detection result set updating means may specifically include the following units.

The detection unit 601: after any monitoring image of the area is obtained, content detection can be performed on the image, and the obtained content detection result is determined as a result to be added.

Matching unit 602: traversing the added results in the current content detection result set to be matched with the results to be added, and judging whether the added results successfully matched with the results to be added exist in the current set; if not, the result to be added can be added into the current set, and the repeated count value of the result is initialized; if so, the repeated count values of the added results which are successfully matched are subjected to increment processing according to a preset increment rule.

Wherein each added result in the set may also have a match failure count value.

The matching unit 602 may also be configured to: if any added result fails to match with the result to be added, performing incremental processing on the matching failure count value of the added result according to a preset incremental rule. The match failure count value may be used at least for: and deleting any added result from the current set after the matching failure count value of the added result is larger than a preset threshold value.

Still further, the matching unit 602 may be specifically configured to: if the result is present, performing increment processing on the repeated count values of all the successfully matched added results according to a preset increment rule, and performing decrement processing on the failed match count values of all the successfully matched added results according to a preset decrement rule.

The matching unit 602 may also be specifically configured to: when the matching failure count value of any added result determined to be the error detection result is larger than a first preset threshold value, the added result is suspected to be deleted from the current set; and deleting any added result which is not determined to be the error detection result from the current set after the matching failure count value of the added result is larger than a second preset threshold value.

Furthermore, each added result in the collection may also have a validity time.

The matching unit 602 may also be configured to: if any added result fails to be matched with the result to be added, the effective time of the added result can be subjected to decrement treatment according to a preset decrement rule; the effective time can be at least used to: when the effective time of any added result is 0, the added result can be deleted from the current set.

Still further, the matching unit 602 may specifically be configured to: if so, the repeated count value and the effective time of the added results which are successfully matched are subjected to increment processing according to a preset increment rule.

Furthermore, each added result in the set that is determined to be a false detection result may also have a vanishing time.

The matching unit 602 may also be configured to: if any added result fails to be matched with the result to be added, incremental processing can be carried out on the vanishing time of the added result according to a preset incremental rule; the vanishing time is at least for: when the vanishing time of any added result is greater than a preset threshold, the added result can be deleted from the current set.

Still further, the matching unit 602 may be specifically configured to: if so, the repeated count values of the added results which are all successfully matched can be subjected to increment processing according to a preset increment rule, and the disappearance time of the added results which are all successfully matched and are determined to be error detection results can be subjected to decrement processing according to a preset decrement rule.

As shown in fig. 7, a schematic structural diagram of an apparatus for checking image content detection results is provided in the present specification. The apparatus may specifically include the following units, and specific steps may be performed based on the set in the apparatus shown in fig. 6.

Monitoring unit 701: and monitoring the repeated count value of the added result in the set.

The judgment unit 702: if the repeated count value of any added result is larger than the preset threshold value, determining that the added result is an error detection result.

The apparatus may further include: determination unit 703: for any to-be-added result, if the added result successfully matched in the set is an error detection result, the to-be-added result is determined to be the error detection result.

The specification also provides a passenger flow statistics device, aiming at the monitoring image of any area, a pedestrian detection result set for checking can be created, and each added result in the set has a repeated count value; the apparatus may include the following 3 units.

First updating unit 801: updating the pedestrian detection result set and determining an error detection result.

The updating may include: after any monitoring image of the area is obtained, pedestrian detection is carried out on the image, and the obtained pedestrian detection result is determined as a result to be added; traversing the added results in the current pedestrian detection result set to be matched with the results to be added, and judging whether the added results successfully matched with the results to be added exist in the current set; if not, adding the result to be added into the current set, and initializing the repeated count value of the result; if so, performing increment processing on repeated count values of all the successfully matched added results according to a preset increment rule; the repetition count is at least for: and when the repeated count value of any added result is larger than a preset threshold value, determining that the added result is an error detection result.

The first pedestrian detection unit 802: after any monitoring image of the area is obtained, pedestrian detection is carried out on the image, and a current pedestrian detection result is obtained.

A statistics unit 803: and determining the error detection result in the current pedestrian detection result according to the error detection results in the set, and counting the number of non-error detection results in the current pedestrian detection result for passenger flow statistics.

The specification also provides a pedestrian attention behavior recognition device, wherein the pedestrian attention behavior at least can comprise commodity taking behavior, commodity trial behavior and try-on behavior; for the monitoring image of any area, a pedestrian detection result set for checking can be created, and each added result in the set has a repeated count value; the device may comprise the following 4 units.

The second updating unit 901: updating the pedestrian detection result set and determining an error detection result.

The second pedestrian detection unit 902: after any monitoring image of the area is obtained, pedestrian detection is carried out on the image, and a current pedestrian detection result is obtained.

The identification unit 903: determining an error detection result contained in the current pedestrian detection result according to the error detection results in the set; aiming at any one of the current pedestrian detection results, whether the behavior of the pedestrian corresponding to the non-false detection result belongs to any pedestrian attention behavior or not is identified by utilizing an image identification technology.

The detailed explanation of the four device embodiments may refer to the method embodiments, and are not repeated herein.

The embodiments of the present disclosure also provide a computer device, which at least includes a memory, a processor, and a computer program stored on the memory and capable of running on the processor, wherein the processor implements an image content detection result set updating method as shown in fig. 3 or an image content detection result checking method as shown in fig. 4 when executing the program.

FIG. 8 is a schematic diagram of a more specific hardware architecture of a computer device according to an embodiment of the present disclosure, where the device may include: a processor 1010, a memory 1020, an input/output interface 1030, a communication interface 1040, and a bus 1050. Wherein processor 1010, memory 1020, input/output interface 1030, and communication interface 1040 implement communication connections therebetween within the device via a bus 1050.

The processor 1010 may be implemented by a general-purpose CPU (Central Processing Unit ), a microprocessor, an Application SPECIFIC INTEGRATED Circuit (ASIC), or one or more integrated circuits, etc. for executing related programs to implement the technical solutions provided in the embodiments of the present disclosure.

The Memory 1020 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory ), static storage, dynamic storage, etc. Memory 1020 may store an operating system and other application programs, and when the embodiments of the present specification are implemented in software or firmware, the associated program code is stored in memory 1020 and executed by processor 1010.

The input/output interface 1030 is used to connect with an input/output module for inputting and outputting information. The input/output module may be configured as a component in a device (not shown) or may be external to the device to provide corresponding functionality. Wherein the input devices may include a keyboard, mouse, touch screen, microphone, various types of sensors, etc., and the output devices may include a display, speaker, vibrator, indicator lights, etc.

Communication interface 1040 is used to connect communication modules (not shown) to enable communication interactions of the present device with other devices. The communication module may implement communication through a wired manner (such as USB, network cable, etc.), or may implement communication through a wireless manner (such as mobile network, WIFI, bluetooth, etc.).

Bus 1050 includes a path for transferring information between components of the device (e.g., processor 1010, memory 1020, input/output interface 1030, and communication interface 1040).

It should be noted that although the above-described device only shows processor 1010, memory 1020, input/output interface 1030, communication interface 1040, and bus 1050, in an implementation, the device may include other components necessary to achieve proper operation. Furthermore, it will be understood by those skilled in the art that the above-described apparatus may include only the components necessary to implement the embodiments of the present description, and not all the components shown in the drawings.

The present embodiment also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements an image content detection result set updating method as shown in fig. 3 or an image content detection result checking method as shown in fig. 4.

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 storage media for a computer 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 Disks (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

From the foregoing description of embodiments, it will be apparent to those skilled in the art that the present embodiments may be implemented in software plus a necessary general purpose hardware platform. Based on such understanding, the technical solutions of the embodiments of the present specification may be embodied in essence or what contributes to the prior art in the form of a software product, which may be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the embodiments or some parts of the embodiments of the present specification.

The system, apparatus, module or unit set forth in the above embodiments may be implemented in particular by a computer chip or entity, or by a product having a certain function. A typical implementation device is a computer, which may be in the form of a personal computer, laptop computer, cellular telephone, camera phone, smart phone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or a combination of any of these devices.

In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for the device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments for relevant points. The apparatus embodiments described above are merely illustrative, in which the modules illustrated as separate components may or may not be physically separate, and the functions of the modules may be implemented in the same piece or pieces of software and/or hardware when implementing the embodiments of the present disclosure. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

The foregoing is merely a specific implementation of the embodiments of this disclosure, and it should be noted that, for a person skilled in the art, several improvements and modifications can be made without departing from the principles of the embodiments of this disclosure, which should also be regarded as protection of the embodiments of this disclosure.

Claims

1. An image content detection result set updating method includes the steps that a content detection result set for checking is established for a monitoring image of any area, and each added result in the set has a repeated count value; the method comprises the following steps:

2. The method of claim 1, each added result in the set further having a match failure count value; if any added result fails to be matched with the result to be added, performing incremental processing on a matching failure count value of the added result according to a preset incremental rule;

The match failure count value is at least for: and deleting any added result from the current set after the matching failure count value of the added result is larger than a preset threshold value.

3. The method of claim 2, wherein if the added result repetition count value of all successful matches is incrementally processed according to a preset increment rule, comprising:

If the result is present, performing increment processing on the repeated count values of all the successfully matched added results according to a preset increment rule, and performing decrement processing on the failed match count values of all the successfully matched added results according to a preset decrement rule.

4. The method of claim 2, wherein deleting the added result from the current set when the match failure count value of any added result is greater than a preset threshold value comprises:

when the matching failure count value of any added result determined to be the error detection result is larger than a first preset threshold value, deleting the added result from the current set;

And deleting any added result which is not determined to be the error detection result from the current set after the matching failure count value of the added result is larger than a second preset threshold value.

5. The method of claim 1, each added result in the set further having a validity time; if any added result fails to be matched with the result to be added, carrying out decrement processing on the effective time of the added result according to a preset decrement rule;

The effective time is at least for: when the effective time of any added result is 0, the added result is deleted from the current set.

6. The method of claim 5, wherein if the added result repetition count value is incremented according to a preset increment rule, wherein the increment processing is performed on the added result repetition count value of all successful matching, and the method comprises:

If so, performing increment processing on the repeated count value and the effective time of the added results which are successfully matched according to a preset increment rule.

7. The method of claim 1, each added result in the set determined to be a false detection result further having a vanishing time; if any added result fails to be matched with the result to be added, performing incremental processing on the vanishing time of the added result according to a preset incremental rule;

the vanishing time is at least for: and deleting any added result from the current set after the vanishing time of the added result is greater than a preset threshold.

8. The method of claim 7, wherein if the added result repetition count value of all successful matches is incrementally processed according to a preset increment rule, comprising:

if so, performing increment processing on the repeated count values of the added results which are all successfully matched according to a preset increment rule, and performing decrement processing on the vanishing time of the added results which are all successfully matched and are determined to be error detection results according to a preset decrement rule.

9. An image content detection result checking method based on the updated set of the image content detection result updating method according to any one of claims 1 to 8, comprising:

10. The method of claim 9, further comprising:

And for any to-be-added result, if the added result successfully matched in the set is an error detection result, determining the to-be-added result as the error detection result.

11. A passenger flow statistical method is characterized in that a pedestrian detection result set for checking is established aiming at a monitoring image of any area, and each added result in the set has a repeated count value; the method comprises the following steps:

Updating the pedestrian detection result set and determining an error detection result;

the updating includes: after any monitoring image of the area is obtained, pedestrian detection is carried out on the image, and the obtained pedestrian detection result is determined as a result to be added; traversing the added results in the current pedestrian detection result set to be matched with the to-be-added results, and judging whether the added results successfully matched with the to-be-added results exist in the current set; if not, adding the result to be added to the current set, and initializing a repetition count value of the result; if so, performing increment processing on repeated count values of all the successfully matched added results according to a preset increment rule; the repetition count value is at least for: when the repeated count value of any added result is larger than a preset threshold value, determining that the added result is an error detection result;

After any monitoring image of the area is obtained, pedestrian detection is carried out on the image, and a current pedestrian detection result is obtained;

And determining the error detection result in the current pedestrian detection result according to the error detection results in the set, and counting the number of non-error detection results in the current pedestrian detection result for passenger flow statistics.

12. The pedestrian attention behavior identification method at least comprises commodity taking behavior, commodity trial behavior and try-on behavior; creating a pedestrian detection result set for checking aiming at the monitoring image of any area, wherein each added result in the set has a repeated count value; the method comprises the following steps:

Determining an error detection result contained in the current pedestrian detection result according to the error detection results in the set; and aiming at any one of the current pedestrian detection results, identifying whether the behavior of the pedestrian corresponding to the non-false detection result belongs to any pedestrian attention behavior or not by utilizing an image identification technology.

13. An image content detection result set updating device creates a content detection result set for checking for a monitoring image of any area, wherein each added result in the set has a repeated count value; the device comprises:

14. An image content detection result checking apparatus based on the updated set of the image content detection result updating apparatus according to claim 13, comprising:

15. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any one of claims 1 to 12 when the program is executed by the processor.