CN116434042A

CN116434042A - Identification quality detection method, device, computer equipment and storage medium

Info

Publication number: CN116434042A
Application number: CN202310161611.3A
Authority: CN
Inventors: 蔡佳; 韦胜钰; 黄林轶; 徐华伟
Original assignee: China Electronic Product Reliability and Environmental Testing Research Institute
Current assignee: China Electronic Product Reliability and Environmental Testing Research Institute
Priority date: 2023-02-23
Filing date: 2023-02-23
Publication date: 2023-07-14

Abstract

The application relates to a recognition quality detection method, a recognition quality detection device, computer equipment and a storage medium, and relates to the technical field of simulation experiments. The method comprises the following steps: constructing an original three-dimensional simulation environment corresponding to the equipment to be detected according to the use scene of the equipment to be detected; determining at least two groups of object attributes of the identification object according to the original three-dimensional simulation environment; wherein the object pose attributes and/or object identity attributes of the different sets of object attributes are different; according to the object attributes of each group, deploying the identified objects in the original three-dimensional simulation environment to obtain a target three-dimensional simulation environment; and carrying out recognition quality detection on the equipment to be detected based on the target original three-dimensional simulation environment to obtain a recognition quality detection result. The accuracy of determining the recognition quality detection result is improved, so that the recognition quality detection result is more in line with the actual application situation of the equipment to be detected.

Description

Identification quality detection method, device, computer equipment and storage medium

Technical Field

The present disclosure relates to the field of simulation experiments, and in particular, to a method and apparatus for detecting recognition quality, a computer device, and a storage medium.

Background

Along with the continuous development of the target recognition technology, the target recognition technology is applied to more and more devices; in order to ensure that the device can continuously and effectively complete the target recognition operation, the device needs to be subjected to recognition quality detection regularly.

In the prior art, an image containing an identification object can be determined according to the identification object corresponding to the equipment; the control equipment performs target recognition operation on the image containing the recognition object, and further determines a recognition quality detection result of the equipment according to the accuracy of the recognition result.

However, when the device is detected by the image containing the identification object, the image containing the identification object has a certain difference from the real target, and the image containing the identification object also lacks depth information from the target, so that the accuracy of the method for detecting the identification quality of the device in the prior art is low, and the real identification quality of the device cannot be effectively obtained.

Disclosure of Invention

In view of the foregoing, it is desirable to provide a recognition quality detecting method, apparatus, computer device, and storage medium capable of improving the accuracy of detection of recognition quality of a device to be detected.

In a first aspect, the present application provides a recognition quality detection method. The method comprises the following steps:

constructing an original three-dimensional simulation environment corresponding to the equipment to be detected according to the use scene of the equipment to be detected;

determining at least two groups of object attributes of the identification object according to the original three-dimensional simulation environment; wherein the object pose attributes and/or object identity attributes of the different sets of object attributes are different;

according to the object attributes of each group, deploying the identified objects in the original three-dimensional simulation environment to obtain a target three-dimensional simulation environment;

and carrying out recognition quality detection on the equipment to be detected based on the target original three-dimensional simulation environment to obtain a recognition quality detection result.

In one embodiment, according to a usage scenario of a device to be detected, an original three-dimensional simulation environment corresponding to the device to be detected is constructed, including:

determining environmental impact factors of equipment to be detected based on the use scene of the equipment to be detected; wherein the environmental impact factors include climate impact factors and/or light impact factors;

determining at least one set of test environment parameters based on the environmental impact factors;

and constructing an original three-dimensional simulation environment corresponding to the equipment to be detected under each group of test environment parameters.

In one embodiment, the method further comprises:

determining a range detection point based on the original three-dimensional simulation environment;

determining the identification range of the equipment to be detected in the original three-dimensional simulation environment according to the identification result of the equipment to be detected on the identification object at the range detection point;

correspondingly, based on the original three-dimensional simulation environment of the target, the identification quality detection of the equipment to be detected comprises the following steps:

and carrying out recognition quality detection on the equipment to be detected based on the target three-dimensional simulation environment in the recognition range.

In one embodiment, after determining the identification range of the device to be detected in the original three-dimensional simulation environment, the method further comprises:

determining a range verification point based on the identification range; wherein the range verification point is different from the range test point in position;

and updating the identification range according to the identification result of the equipment to be detected on the identification object at the range verification point.

In one embodiment, deploying an identified object in an original three-dimensional simulated environment based on sets of object properties includes:

according to the determined group number of the object attributes, determining the deployment position of the identification object corresponding to each group of the object attributes in the original three-dimensional simulation environment;

Based on the deployment position, deploying the identification object corresponding to each object attribute in the original three-dimensional simulation environment.

In one embodiment, performing recognition quality detection on a device to be detected to obtain a recognition quality detection result includes:

performing identification quality detection on equipment to be detected, and determining an identification accuracy index and a response time index;

and determining a recognition quality detection result according to the recognition accuracy index and the response time index.

In a second aspect, the present application also provides an identification quality detection apparatus. The device comprises:

the construction module is used for constructing an original three-dimensional simulation environment corresponding to the equipment to be detected according to the use scene of the equipment to be detected;

the determining module is used for determining at least two groups of object attributes of the identification object according to the original three-dimensional simulation environment; wherein the object pose attributes and/or object identity attributes of the different sets of object attributes are different;

the deployment module is used for deploying the identified objects in the original three-dimensional simulation environment according to the object attributes of each group to obtain a target three-dimensional simulation environment;

the detection module is used for carrying out recognition quality detection on the equipment to be detected based on the original three-dimensional simulation environment of the target to obtain a recognition quality detection result.

In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the identification quality detection method according to any of the embodiments of the first aspect described above when the processor executes the computer program.

In a fourth aspect, the present application also provides a computer-readable storage medium. A computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements an identification quality detection method as in any of the embodiments of the first aspect described above.

In a fifth aspect, the present application also provides a computer program product. A computer program product comprising a computer program which, when executed by a processor, implements an identification quality detection method as in any of the embodiments of the first aspect described above.

According to the technical scheme, the original three-dimensional simulation environment is constructed according to the use scene of the equipment to be detected, so that the constructed simulation environment is more in line with the actual use scene, at least two groups of object attributes of the object are identified according to the original three-dimensional simulation environment, the diversity of the identified object in the original three-dimensional simulation environment is improved, and the determined object attributes are ensured to be more matched with the actual use scene. After the identification objects are arranged in the original three-dimensional simulation environment according to the determined object attributes, the identification quality detection is carried out on the equipment to be detected, and as the three-dimensional simulation environment at the moment contains X, Y and Z three-dimensional environment information and contains the multi-attribute identification objects, compared with the two-dimensional image containing a single identification object in the prior art, the identification quality detection is carried out on the equipment to be detected, the identification quality detection result can be obtained in the simulated real environment, the difference between the identification objects and the real targets is reduced, and the real identification quality of the equipment is effectively obtained.

Drawings

Fig. 1 is an application environment diagram of an identification quality detection method provided in an embodiment of the present application;

fig. 2 is a flowchart of an identification quality detection method according to an embodiment of the present application;

FIG. 3 is a flowchart illustrating steps for constructing an original three-dimensional simulation environment according to an embodiment of the present application;

FIG. 4 is a flowchart illustrating steps for determining an identification range according to an embodiment of the present application;

FIG. 5 is an exemplary diagram for determining an identification range according to an embodiment of the present application;

FIG. 6 is a flowchart of steps for deploying an identified object in an original three-dimensional simulated environment, provided in an embodiment of the present application;

fig. 7 is a schematic diagram of a space rectangular coordinate system with a device to be detected as an origin in an embodiment of the present application;

FIG. 8 is a flowchart illustrating steps for determining recognition quality detection results according to an embodiment of the present application;

FIG. 9 is a flowchart illustrating steps of another method for detecting recognition quality according to an embodiment of the present application;

FIG. 10 is a diagram illustrating an exemplary three-dimensional simulation environment for a target according to an embodiment of the present application;

FIG. 11 is a block diagram of a first recognition quality detecting apparatus according to an embodiment of the present application;

FIG. 12 is a block diagram of a second recognition quality detecting apparatus according to an embodiment of the present application;

Fig. 13 is a block diagram of a third recognition quality detecting apparatus according to an embodiment of the present application;

fig. 14 is a block diagram of a fourth recognition quality detecting apparatus according to an embodiment of the present application;

fig. 15 is a block diagram of a fifth recognition quality detecting apparatus according to an embodiment of the present application;

fig. 16 is a block diagram of a sixth recognition quality detecting apparatus according to an embodiment of the present application;

fig. 17 is an internal structural view of a computer device in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.

It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application. In the description of the present application, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.

In the prior art, a two-dimensional image or video stream containing an identification object is mainly adopted as a test set to perform identification quality detection, however, because a certain difference exists between the image containing the identification object and a real target, and depth information from the target is also lacking in the image containing the identification object, the accuracy of the method for performing identification quality detection on equipment in the prior art is low, and the real identification quality of the equipment cannot be effectively obtained.

In addition, the method for detecting the recognition quality of the equipment in the prior art lacks a detection process aiming at the shooting performance of the equipment, namely, the situation that the shot image is subjected to overexposure, shooting deformity, shooting distortion and the like possibly occurs in the process of detecting the recognition quality of the equipment, and the accuracy of equipment target recognition can be influenced by the image with the situation.

Based on the above situation, the recognition quality detection method provided in the embodiment of the present application may be applied to an application environment as shown in fig. 1. In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in FIG. 1. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing acquired data identifying the quality detection method. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a recognition quality detection method.

The application discloses a recognition quality detection method, a recognition quality detection device, computer equipment and a storage medium, wherein the computer equipment constructs an original three-dimensional simulation environment according to a use scene of equipment to be detected; determining at least two groups of object attributes according to the original three-dimensional simulation environment; according to the object attributes of each group, the identified objects are deployed in the original three-dimensional simulation environment, so that a target three-dimensional simulation environment is obtained; and carrying out recognition quality detection on the equipment to be detected according to the target three-dimensional simulation environment to obtain a recognition quality detection result.

In one embodiment, as shown in fig. 2, fig. 2 is a flowchart of an identification quality detection method provided in an embodiment of the present application, and the identification quality detection method performed by the computer device in fig. 1 may include the following steps:

step 201, constructing an original three-dimensional simulation environment corresponding to the equipment to be detected according to the use scene of the equipment to be detected.

The usage scenario of the device to be detected may include, but is not limited to: indoor, outdoor, high altitude, woody, underwater, etc.

It should be noted that, the usage scenario of the device to be detected may be a single scenario or a plurality of scenarios, and if the usage scenario of the device to be detected is a single scenario, the device to be detected corresponds to a unique original three-dimensional simulation environment; if the usage scene of the device to be detected is a plurality of scenes, each usage scene of the device to be detected corresponds to an original three-dimensional simulation environment.

In one embodiment of the application, after determining the usage scenario of the device to be detected, determining environmental impact factors to be born by the device to be detected under the usage scenario according to the usage scenario of the device to be detected; setting corresponding environment influence facilities according to the environment influence factors, so as to construct an original three-dimensional simulation environment; after all environmental influence factors required to be born by the equipment to be detected can be simulated according to the environmental influence facilities, the construction of the original three-dimensional simulation environment is completed, so that the constructed original three-dimensional simulation environment and the use scene of the equipment to be detected are ensured to have enough similarity.

Among other things, environmental impact facilities may include, but are not limited to: fluorescent lamp, rainwater spray device and fog simulation device. Environmental impact factors may include, but are not limited to: the light intensity, temperature, rainfall, mist reduction, etc. are not limited here by the environmental impact facilities and the types of environmental impact factors.

For example, when setting a responsive environmental impact facility according to environmental impact factors, the following may be included: if the environmental impact factors comprise illumination intensity, the illumination intensity born by the equipment to be detected in the use scene can be simulated through the equipment fluorescent lamp; if the environmental impact factors comprise rainfall, simulating rainfall conditions born by the equipment to be detected in a use scene through a rainwater spraying device; if the environmental influence factors comprise fog reduction, the fog reduction condition born by the equipment to be detected in the use scene can be simulated through the fog simulation device.

Step 202, determining at least two groups of object attributes of the identified objects according to the original three-dimensional simulation environment.

The object attribute of the identification object comprises an object posture attribute representing the corresponding posture of the identification object and an object identity attribute representing the identity characteristic of the identification object; and, object pose attributes and/or object identity attributes of different sets of object attributes are different.

Further, taking the example of recognizing the object as a face, the object gesture attribute may include, but is not limited to: the face orientation of the recognition object, the facial expression of the recognition object, and the like. Object identity attributes may include, but are not limited to: the young people wait for the identification of objects, the old people for the identification of objects, the long-term identification of objects, the short-term identification of objects, etc.

It should be noted that, in order to ensure that the actual condition of the device to be detected can be reflected by the subsequent determination of the recognition quality detection result of the device to be detected, sufficient acquisition of the object attributes should be ensured when determining at least two sets of object attributes of the recognition object.

In one embodiment of the present application, when at least two sets of object attributes of an identified object need to be determined, all object identity attributes possibly included in the identified object and all object gesture attributes possibly included in the identified object may be determined according to the identified object targeted by the device to be detected, and all object identity attributes possibly included in the identified object and all object gesture attributes possibly included in the identified object are combined in pairs to obtain at least two sets of object attributes.

And 203, according to the object attributes of each group, deploying the identified objects in the original three-dimensional simulation environment to obtain a target three-dimensional simulation environment.

When the identification object needs to be deployed in the original three-dimensional simulation environment, the identification range of the device to be detected in the original three-dimensional simulation environment can be predetermined, the deployment position corresponding to each group of object attributes is determined in the identification range, and the identification object corresponding to each group of object attributes is deployed in the corresponding deployment position, so that the target three-dimensional simulation environment is obtained.

In one embodiment of the present application, when the deployment location corresponding to each set of object attributes needs to be determined within the recognition range, the following two cases may be specifically included: in the first case, a deployment position corresponding to each group of object attributes is randomly determined in an identification range; in the second case, based on a preset deployment rule, the deployment position corresponding to each set of object attributes is determined in the recognition range.

Further illustratively, the pre-set deployment rules may include, but are not limited to: at least one deployment position is arranged at the corner of the identification range, at least one deployment position is arranged at the farthest position from the equipment to be detected in the identification range, and at least one deployment position is arranged at the nearest position from the equipment to be detected in the identification range. It will be appreciated that there are many kinds of deployment rules, and the kinds of deployment rules are not limited here.

And 204, performing recognition quality detection on the equipment to be detected based on the target original three-dimensional simulation environment to obtain a recognition quality detection result.

It should be noted that, when the device to be detected performs recognition quality detection, the accuracy index of the device to be detected for the recognition object and the response time of the device to be detected for performing target recognition can be determined according to the recognition result of the device to be detected for the recognition object in the original three-dimensional simulation environment. And determining a quality detection result according to the accuracy index and the response time.

Further, the accuracy index of the equipment to be detected for the identification objects can be determined according to the number of the identification objects identified by the equipment to be detected in the target original three-dimensional simulation environment and the number of the identification objects actually deployed in the target original three-dimensional simulation environment.

Further, the response time of the device to be detected for target recognition can be determined according to the time when the device to be detected outputs the recognition result and the time when the device to be detected starts to operate.

According to the identification quality detection method, the original three-dimensional simulation environment is constructed according to the use scene of the equipment to be detected, so that the constructed simulation environment is more in line with the actual use scene, at least two groups of object attributes of the object are identified according to the original three-dimensional simulation environment, the diversity of the identified object in the original three-dimensional simulation environment is improved, and the determined object attributes are ensured to be more matched with the actual use scene. After the identification objects are arranged in the original three-dimensional simulation environment according to the determined object attributes, the identification quality detection is carried out on the equipment to be detected, and as the three-dimensional simulation environment at the moment contains X, Y and Z three-dimensional environment information and contains the multi-attribute identification objects, compared with the two-dimensional image containing a single identification object in the prior art, the identification quality detection is carried out on the equipment to be detected, the identification quality detection result can be obtained in the simulated real environment, the difference between the identification objects and the real targets is reduced, and the real identification quality of the equipment is effectively obtained.

It should be noted that, by determining environmental influence factors, an original three-dimensional simulation environment can be constructed according to the environmental influence factors; optionally, as shown in fig. 3, fig. 3 is a flowchart of a step of constructing an original three-dimensional simulation environment according to an embodiment of the present application, and specifically, the step of constructing the original three-dimensional simulation environment may include the following steps:

step 301, determining environmental impact factors of the device to be detected based on the usage scenario of the device to be detected.

Wherein the environmental impact factors include climate impact factors and/or light impact factors. By way of example, climate influencing factors may include, but are not limited to: rainfall, snowfall, mist reduction and other factors; light influencing factors may include, but are not limited to: illumination intensity, illumination time light source direction, etc.

It should be noted that, because the environmental impact factors corresponding to different usage scenarios of the device to be detected are also different, the environmental impact factors corresponding to the device to be detected need to be determined according to the usage scenario of the device to be detected, and further, the method for determining the environmental impact factors may include: according to the historical working experience of the staff, determining environmental influence factors corresponding to the use scene of the equipment to be detected; or determining environmental impact factors corresponding to the use scene of the equipment to be detected according to the historical identification quality detection record, and describing the two methods for determining the environmental impact factors in detail:

In an embodiment of the present application, when an environmental impact factor needs to be determined, according to a historical working experience of a worker, an impact factor corresponding to a usage scenario of the device to be detected in a past working experience may be determined, so as to determine the environmental impact factor of the device to be detected.

In another embodiment of the present application, when an environmental impact factor needs to be determined, a history detection device that is the same as a usage scenario of a device to be detected may be determined according to a history recognition quality detection record; and determining a history influence factor corresponding to the history detection device according to the history recognition quality detection record, wherein the history influence factor corresponding to the history detection device is the environment influence factor corresponding to the equipment to be detected because the use scene of the history detection device is the same as that of the equipment to be detected.

At step 302, at least one set of test environment parameters is determined based on the environmental impact.

It should be noted that, to ensure that an accurate recognition quality detection result can be obtained according to the original three-dimensional simulation environment, the test environment parameters in various situations can be determined based on the environmental influence factors, for example: if the environmental impact is mist reduction, the test environmental parameters in various cases may include: mist reduction with visibility of 1 meter, mist reduction with visibility of five meters and mist reduction with visibility of ten meters.

Further, when the test environment parameters need to be determined, the intensity of the environmental impact factors can be adjusted, for example, the rainfall amount is adjusted or the snowfall amount is adjusted, the environmental impact factors with different intensities correspond to different test environment parameters, and at least one group of test environment parameters corresponding to the environmental impact factors can be determined according to the environmental impact factors with different intensities.

In an embodiment of the present application, if the environmental impact factors of the device to be detected are not unique, determining the test environmental parameters corresponding to each environmental impact factor, and combining the test environmental parameters of different environmental impact factors with each other to obtain at least one set of combined test environmental parameters, where each test environmental parameter is different.

For example, if the device to be detected includes two environmental impact factors, the two environmental impact factors are rainfall and mist reduction; the rainfall corresponding test environment parameters comprise: the method comprises the steps of carrying out combination of two pairs of testing environment parameters of influence factors by two pairs of the testing environment parameters to obtain six groups of testing environment parameters, wherein the testing environment parameters corresponding to the fog reduction comprise visibility D and visibility E, and the six groups of testing environment parameters comprise the following percentages: 1. rainfall A and visibility D; 2. rainfall A and visibility E; 3. rainfall B and visibility D; 4. rainfall B and visibility E; 5. rainfall C and visibility D; 6. rainfall C and visibility E.

And 303, constructing an original three-dimensional simulation environment corresponding to the equipment to be detected under each set of test environment parameters.

It should be noted that, after determining at least one set of test environment parameters, setting and adjusting corresponding environmental impact facilities according to each test environment parameter, so as to realize setting of environmental impact factors according to the environmental impact facilities.

For example, if a set of test environmental parameters of the equipment to be detected is rainfall A of rainfall and visibility D of the reduced fog, the rainfall born by the equipment to be detected can be simulated by arranging the rainwater spraying device, and the rainwater spraying amount of the rainwater spraying device is regulated, so that the rainwater spraying amount of the rainwater spraying device is the rainfall A; and through setting up fog analogue means simulation and waiting to wait to detect the fog that waiting to detect that equipment born receives, adjust fog analogue means's fog emission for fog analogue means's fog emission can satisfy visibility D.

According to the identification quality detection method, the data base is provided for the subsequent construction of the original three-dimensional simulation environment by determining the environmental influence factors of the equipment to be detected, so that the original three-dimensional simulation environment can be ensured to accord with the actual use scene of the equipment to be detected, and the accuracy of the subsequent identification quality detection result is improved; through at least one group of test environment parameters, the recognition condition of the equipment to be detected under various test environment parameters can be fully considered when the equipment to be detected is subjected to recognition quality detection, the accuracy of the follow-up determination recognition quality detection result is further improved, and the original three-dimensional simulation environment is more in line with the actual use scene of the equipment to be detected.

It should be noted that, the identification range of the device to be detected in the original three-dimensional simulation environment can be determined by determining the range detection point and then according to the range detection point; optionally, as shown in fig. 4, fig. 4 is a flowchart of a step of determining an identification range provided in an embodiment of the present application, and specifically, determining the identification range may include the following steps:

and step 401, constructing an original three-dimensional simulation environment corresponding to the equipment to be detected according to the use scene of the equipment to be detected.

Step 402, determining a range detection point based on the original three-dimensional simulation environment.

When the range detection point needs to be determined, the estimated recognition range of the device to be detected can be determined in advance, and then the range detection point can be determined according to the estimated recognition range of the device to be detected. Specifically, if the estimated recognition range of the device to be detected is known, a range detection point is set outside the edge line of the estimated recognition range, so as to verify whether the estimated recognition range needs to be updated.

Further, if the estimated recognition range of the device to be detected cannot be obtained, a point can be selected as a range detection point along the shooting direction of the device to be detected, and further, whether the device to be detected can perform target recognition at the range detection point is judged, so that the determination of the subsequent recognition range is realized.

Step 403, determining the identification range of the device to be detected in the original three-dimensional simulation environment according to the identification result of the device to be detected on the identification object at the range detection point.

It should be noted that, according to the recognition result of the device to be detected on the recognition object at the range detection point, determining whether the range detection point belongs to the recognition range of the device to be detected; specifically, if the equipment to be detected can identify the identification object at the range detection point, determining that the range detection point belongs to the identification range of the equipment to be detected; if the equipment to be detected cannot identify the identification object at the range detection point, determining that the range detection point does not belong to the identification range of the equipment to be detected.

In one embodiment of the present application, if a pre-estimated recognition range of the device to be detected is predetermined and the range detection point is outside an edge line of the pre-estimated recognition range, when the device to be detected can recognize the recognition object at the range detection point, dividing the range detection point into the recognition range of the device to be detected; when the device to be detected cannot recognize the recognition object at the range detection point, the range detection point is not divided into the recognition range of the device to be detected.

In one embodiment of the present application, if the estimated recognition range of the device to be detected cannot be obtained, when the device to be detected can recognize the recognition object at the range detection point, the range detection point is used as the recognition range of the device to be detected; when the device to be detected cannot recognize the recognition object at the range detection point, it is determined that the range detection point cannot be used as the recognition range of the device to be detected.

Further, when the identification range of the device to be detected is determined according to the identification result of the device to be detected on the object identified at the range detection point, a new range detection point can be selected outside the identification range again, and the operation of determining the identification range of the device to be detected according to the identification result of the device to be detected on the object identified at the range detection point is performed again until the area of the determined identification range reaches the maximum value. As shown in fig. 5, fig. 5 is an exemplary diagram for determining an identification range provided in the embodiment of the present application, where A, B and C in fig. 5 are range detection points selected for determining an identification range of a device to be detected, and a dotted line area in fig. 5 is the identification range of the device to be detected.

It should be noted that, after determining the identification range of the device to be detected in the original three-dimensional simulation environment, the method further includes the following steps: determining a range verification point based on the identification range; wherein the range verification point is different from the range test point in position; and updating the identification range according to the identification result of the equipment to be detected on the identification object at the range verification point.

As an implementation manner, when the identification range needs to be verified, five range verification points can be selected at will in the identification range, target identification is carried out on the range verification points according to equipment to be detected, so that the identification result of the range verification points is obtained, and if all the five range verification points are identified, the determined identification range is accurate, and the identification range does not need to be updated; if at least one range verification point in the five range verification points is not recognized, the determined recognition range is inaccurate, and the recognition range needs to be updated.

As another implementation manner, five range verification points can be arbitrarily selected outside the identification range, target identification is performed on the range verification points according to the equipment to be detected, the identification result of the range verification points is obtained, if none of the five range verification points is identified, the determined identification range is accurate, and the identification range does not need to be updated; if at least one range verification point in the five range verification points is identified, the determined identification range is inaccurate, and the identification range needs to be updated.

As a further implementation manner, five range verification points are arbitrarily selected, the range verification points can be outside the identification range and can be in the identification range, target identification is carried out on the range verification points according to equipment to be detected, the identification result of the range verification points is obtained, and if all the range verification points in the identification range are identified, and none of the range verification points outside the identification range are identified, the determined identification range is accurate, and the identification range is not required to be updated; if at least one range verification point exists in the range verification points in the identification range and is not identified, and/or if at least one range verification point exists in the range verification points outside the identification range and is identified, the determined identification range is inaccurate, and the identification range needs to be updated.

Step 404, determining at least two groups of object attributes of the identified objects according to the original three-dimensional simulation environment; wherein the object pose properties and/or object identity properties of the different sets of object properties are different.

And step 405, according to the object attributes of each group, deploying the identified objects in the original three-dimensional simulation environment to obtain a target three-dimensional simulation environment.

And 406, performing recognition quality detection on the equipment to be detected based on the target three-dimensional simulation environment in the recognition range to obtain a recognition quality detection result.

According to the recognition quality detection method, the recognition range of the equipment to be detected in the original three-dimensional simulation environment is determined by setting the range detection point and according to the range detection point, so that the recognition quality detection result can be successfully obtained later, and the accuracy of the recognition quality detection result is ensured; by setting the range verification point, the verification operation of the identification range is realized, the accuracy of the identification range is improved, the accuracy of the identification quality detection result is further improved, and the influence on the identification quality detection process of the equipment to be detected due to the error of the identification range is prevented.

It should be noted that, the deployment of the identified object in the original three-dimensional simulation environment can be achieved by determining the deployment position; optionally, as shown in fig. 6, fig. 6 is a flowchart of a step of deploying an identified object in an original three-dimensional simulation environment according to an embodiment of the present application, and specifically, deploying the identified object in the original three-dimensional simulation environment may include the following steps:

And step 601, determining the deployment position of the identification object corresponding to each group of object attributes in the original three-dimensional simulation environment according to the determined group number of the object attributes.

It should be noted that there are many methods for determining the deployment position in the original three-dimensional simulation environment, for example, the recognition object may be deployed randomly in the original three-dimensional simulation environment, and the recognition object may be deployed according to the usage scenario of the device to be detected; in summary, there are many methods for determining the deployment location in the original three-dimensional simulation environment, which will not be described in detail herein, and the following details will be described for the two methods for determining the deployment location in the original three-dimensional simulation environment.

In one embodiment of the present application, when randomly deploying an identified object in an original three-dimensional simulated environment, the method may comprise the steps of: according to the number of groups of object attributes, determining the number of identification objects to be deployed in an original three-dimensional simulation environment, deploying the number of identification objects according to the need, and randomly selecting a corresponding number of points in the original three-dimensional simulation environment as deployment positions of the identification objects.

Specifically, as shown in fig. 7, when a corresponding number of points are arbitrarily selected as deployment positions in the original three-dimensional simulation environment, a space rectangular coordinate system can be established by using the device to be detected as an origin, and then the three-dimensional coordinates of each point are determined, wherein the three-dimensional coordinates are the deployment positions of the recognition objects.

In another embodiment of the present application, when the identification object is deployed according to the usage scenario of the device to be detected, the method may include the following steps: according to the use scene of the equipment to be detected, determining a partial area with higher probability of identifying objects in the original three-dimensional simulation environment; according to the number of groups of object attributes, determining the number of identification objects to be deployed in an original three-dimensional simulation environment, deploying the number of identification objects according to the need, and randomly selecting a corresponding number of points in a partial area with high probability of the identification objects in the original three-dimensional simulation environment as deployment positions.

For example, if the usage scene of the device to be detected is traffic monitoring, a partial area with higher probability of occurrence of the recognition object in the original three-dimensional simulation environment is near the sidewalk; according to the number of groups of object attributes, determining the number of identification objects to be deployed in an original three-dimensional simulation environment, deploying the number of identification objects according to the need, and randomly selecting a corresponding number of points near a sidewalk in the original three-dimensional simulation environment as deployment positions.

Step 602, based on the deployment position, deploying the identified object corresponding to each object attribute in the original three-dimensional simulation environment.

When the identified objects need to be deployed in the original three-dimensional simulation environment, the identified objects corresponding to the object attributes are predetermined, and then the identified objects corresponding to the object attributes are randomly placed in the deployment position of the original three-dimensional simulation environment, so that the identified objects corresponding to the object attributes are deployed in the original three-dimensional simulation environment.

According to the recognition quality detection method, the deployment position is determined, so that the recognition object can be deployed in the original three-dimensional simulation environment according to the deployment position, the subsequent flow is ensured to be smoothly carried out, and the subsequent recognition quality detection result can be ensured to be smoothly obtained.

It should be noted that, the recognition quality detection result can be determined by the recognition accuracy index and the response time index; optionally, as shown in fig. 8, fig. 8 is a flowchart of a step of determining an identification quality detection result provided in an embodiment of the present application, and specifically, determining the identification quality detection result may include the following steps:

step 801, performing identification quality detection on equipment to be detected, and determining an identification accuracy index and a response time index.

It should be noted that the identification accuracy index may include at least one of a detection rate of the device to be detected, an accuracy rate of the device to be detected, and a false detection rate of the device to be detected. The detection rate of the equipment to be detected is used for representing the ratio of the number of the identification objects identified by the equipment to be detected to the number of the identification objects actually deployed in the target original three-dimensional simulation environment; the accuracy of the equipment to be detected is used for representing the ratio of the number of the identification objects correctly identified by the equipment to be detected to the number of the identification objects identified by the equipment to be detected in the target original three-dimensional simulation environment; the false detection rate of the device to be detected is used for representing the ratio of the number of the recognition objects which are recognized by the device to be detected in error to the number of the recognition objects actually deployed in the target original three-dimensional simulation environment.

Further, the detection rate of the equipment to be detected can be determined according to the number of the identified objects identified by the equipment to be detected in the target original three-dimensional simulation environment and the number of the identified objects actually deployed in the target original three-dimensional simulation environment;

in summary, the calculation formula (1) of the detection rate of the device to be detected can be expressed as follows:

further, the accuracy of the equipment to be detected can be determined according to the number of the identification objects which are correctly identified by the equipment to be detected in the target original three-dimensional simulation environment and the number of the identification objects which are identified by the equipment to be detected in the target original three-dimensional simulation environment;

in summary, the calculation formula (2) of the accuracy of the device to be detected can be expressed as follows:

further, the false detection rate of the equipment to be detected can be determined according to the number of recognition objects which are erroneously recognized in the target original three-dimensional simulation environment by the equipment to be detected and the number of recognition objects which are actually deployed in the target original three-dimensional simulation environment;

in summary, the calculation formula (3) of the false detection rate of the device to be detected can be expressed as follows:

the response time index is used to represent the time required for the device to be detected to perform the target recognition from the beginning to output the recognition result.

In summary, the calculation formula (4) of the response time index can be expressed as follows:

response time index=time when the device to be detected outputs the target recognition result-time … … when the device to be detected starts target recognition (4)

Step 802, determining a recognition quality detection result according to the recognition accuracy index and the response time index.

The recognition quality detection result is a generic term for a recognition accuracy index and a response time index.

Further, since the identification accuracy index may include at least one of a detection rate of the device to be detected, an accuracy rate of the device to be detected, and a false detection rate of the device to be detected, if the identification accuracy index includes the detection rate of the device to be detected and the accuracy rate of the device to be detected, the identification quality detection result includes: response time index, detection rate of equipment to be detected and accuracy rate of equipment to be detected.

According to the identification quality detection method, through determining the identification accuracy index and the response time index, the identification quality detection result is ensured to fully embody the actual condition of the equipment to be detected, and the identification quality of the equipment to be detected is determined according to the identification quality detection result.

In an embodiment of the present application, as shown in fig. 9, fig. 9 is a flowchart of steps of another method for detecting recognition quality provided in the embodiment of the present application, when a device to be detected needs to be detected for detection of recognition quality, the method specifically may include the following steps:

step 901, determining environmental impact factors of the device to be detected based on the usage scenario of the device to be detected.

At step 902, at least one set of test environment parameters is determined based on the environmental impact.

And 903, constructing an original three-dimensional simulation environment corresponding to the equipment to be detected under each set of test environment parameters.

In step 904, a range detection point is determined based on the original three-dimensional simulated environment.

Step 905, determining the identification range of the device to be detected in the original three-dimensional simulation environment according to the identification result of the device to be detected on the identification object at the range detection point.

Step 906, determining a range verification point based on the identification range; wherein the range verification point is located differently than the range test point.

And step 907, updating the identification range according to the identification result of the equipment to be detected on the identification object at the range verification point.

Step 908, determining at least two sets of object properties of the identified object based on the original three-dimensional simulated environment.

Step 909, determining the deployment position of the identified object corresponding to each group of object attributes in the original three-dimensional simulation environment according to the determined group number of the object attributes.

Step 910, based on the deployment location, deploying the identified object corresponding to each object attribute in the original three-dimensional simulation environment, so as to obtain the target three-dimensional simulation environment.

For example, if the environmental impact factors of the to-be-detected device are respectively: rainfall, fog reduction and illumination; and a group of test environment parameters of the equipment to be detected are rainfall A of rainfall, visibility D of defogging and illumination with illumination intensity F; therefore, the rainfall born by the equipment to be detected can be simulated by arranging the rainwater spraying device, and the rainwater spraying amount of the rainwater spraying device is adjusted, so that the rainwater spraying amount of the rainwater spraying device is the rainfall A; the fog reduction born by the equipment to be detected is simulated by arranging the fog simulation device, and the fog discharge amount of the fog simulation device is regulated, so that the fog discharge amount of the fog simulation device can meet the visibility D; and the illumination born by the equipment to be detected is simulated by arranging the fluorescent lamp, so that the illumination intensity of the fluorescent lamp is adjusted, and the illumination intensity of the fluorescent lamp can meet the requirement that the illumination intensity is F. According to the simulation of the rainwater spraying device, the fog simulation device and the fluorescent lamp, an original three-dimensional simulation environment is obtained, after the identification range of the original three-dimensional simulation environment is determined, an identification object is arranged in the identification range of the original three-dimensional simulation environment, and a target three-dimensional simulation environment is obtained. The three-dimensional simulation environment of the target to be detected may be shown in fig. 10, and fig. 10 is an exemplary diagram of the three-dimensional simulation environment of the target provided in the embodiment of the present application.

Step 911, based on the target three-dimensional simulation environment in the identification range, the identification quality detection is performed on the equipment to be detected, and the identification accuracy index and the response time index are determined.

Step 912, determining the recognition quality detection result according to the recognition accuracy index and the response time index.

It should be understood that, although the steps in the flowcharts related to the above embodiments are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, the embodiment of the application also provides an identification quality detection device for realizing the identification quality detection method. The implementation of the solution provided by the device is similar to the implementation described in the above method, so the specific limitation in one or more embodiments of the identification quality detection device provided below may refer to the limitation of the identification quality detection method hereinabove, and will not be repeated herein.

In one embodiment, as shown in fig. 11, fig. 11 is a block diagram of a first recognition quality detecting apparatus according to an embodiment of the present application, and provides a recognition quality detecting apparatus, including: a construction module 10, a first determination module 20, a deployment module 30, and a detection module 40, wherein:

the construction module 10 is configured to construct an original three-dimensional simulation environment corresponding to the device to be detected according to a usage scenario of the device to be detected.

A first determining module 20, configured to determine at least two sets of object attributes of the identified object according to the original three-dimensional simulation environment; wherein the object pose properties and/or object identity properties of the different sets of object properties are different.

The deployment module 30 is configured to deploy the identified object in the original three-dimensional simulation environment according to the object attributes of each group, so as to obtain a target three-dimensional simulation environment.

The detection module 40 is configured to perform recognition quality detection on the device to be detected based on the target original three-dimensional simulation environment, so as to obtain a recognition quality detection result.

According to the recognition quality detection device, the original three-dimensional simulation environment is constructed according to the use scene of the equipment to be detected, so that the constructed simulation environment is more in line with the actual use scene, at least two groups of object attributes of the object are recognized according to the original three-dimensional simulation environment, the diversity of the recognized object in the original three-dimensional simulation environment is improved, and the determined object attributes are ensured to be more matched with the actual use scene. After the identification objects are arranged in the original three-dimensional simulation environment according to the determined object attributes, the identification quality detection is carried out on the equipment to be detected, and as the three-dimensional simulation environment at the moment contains X, Y and Z three-dimensional environment information and contains the multi-attribute identification objects, compared with the two-dimensional image containing a single identification object in the prior art, the identification quality detection is carried out on the equipment to be detected, the identification quality detection result can be obtained in the simulated real environment, the difference between the identification objects and the real targets is reduced, and the real identification quality of the equipment is effectively obtained.

In one embodiment, as shown in fig. 12, fig. 12 is a block diagram of a second recognition quality detecting apparatus according to an embodiment of the present application, and a recognition quality detecting apparatus is provided, where a building module 10 includes: a first determination unit 11, a second determination unit 12, and a construction unit 13, wherein:

a first determining unit 11, configured to determine an environmental impact factor of the device to be detected based on a usage scenario of the device to be detected; wherein the environmental impact factors include climate impact factors and/or light impact factors.

A second determining unit 12 for determining at least one set of test environment parameters based on the environmental impact.

And the construction unit 13 is used for constructing an original three-dimensional simulation environment corresponding to the equipment to be detected under each set of test environment parameters.

In one embodiment, as shown in fig. 13, fig. 13 is a block diagram of a third recognition quality detecting device according to an embodiment of the present application, and a recognition quality detecting device is provided, where the recognition quality detecting device further includes: a second determination module 50 and a third determination module 60, wherein:

a second determining module 50 is configured to determine a range detection point based on the original three-dimensional simulation environment.

A third determining module 60, configured to determine, according to a recognition result of the device to be detected on the object to be recognized at the range detection point, a recognition range of the device to be detected in the original three-dimensional simulation environment;

correspondingly, the detection module 40 is also used for detecting the identification quality of the equipment to be detected based on the target three-dimensional simulation environment in the identification range.

In one embodiment, as shown in fig. 14, fig. 14 is a block diagram of a fourth recognition quality detecting device according to an embodiment of the present application, and a recognition quality detecting device is provided, where the recognition quality detecting device further includes: a fourth determination module 70 and an update module 80, wherein:

a fourth determination module 70 for determining a range verification point based on the identification range; wherein the range verification point is different from the range test point in position;

and the updating module 80 is configured to update the identification range according to the identification result of the device to be detected on the identification object at the range verification point.

In one embodiment, as shown in fig. 15, fig. 15 is a block diagram of a fifth recognition quality detecting apparatus according to an embodiment of the present application, and a recognition quality detecting apparatus is provided, where a deployment module 30 includes: a third determination unit 31 and a deployment unit 32, wherein:

The third determining unit 31 is configured to determine, according to the determined number of groups of object attributes, a deployment position of an identified object corresponding to each group of object attributes in the original three-dimensional simulation environment.

The deployment unit 32 is configured to deploy, based on the deployment location, the identified object corresponding to each object attribute in the original three-dimensional simulation environment.

In one embodiment, as shown in fig. 16, fig. 16 is a block diagram of a sixth recognition quality detecting apparatus according to an embodiment of the present application, and a recognition quality detecting apparatus is provided, in which a detecting module 40 includes: a fourth determination unit 41 and a fifth determination unit 42, wherein:

a fourth determining unit 41, configured to perform recognition quality detection on the device to be detected, and determine a recognition accuracy index and a response time index.

A fifth determining unit 42 for determining the recognition quality detection result according to the recognition accuracy index and the response time index.

The above-described respective modules in the recognition quality detecting apparatus may be implemented in whole or in part by software, hardware, or a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided, which may be a terminal, and the internal structure thereof may be as shown in fig. 17. The computer device includes a processor, a memory, an input/output interface, a communication interface, a display unit, and an input means. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a recognition quality detection method. The display unit of the computer device is used for forming a visual picture, and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be a key, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.

It will be appreciated by those skilled in the art that the structure shown in fig. 17 is merely a block diagram of a portion of the structure associated with the present application and is not limiting of the computer device to which the present application applies, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:

In one embodiment, the processor when executing the computer program further performs the steps of:

In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:

In one embodiment, the computer program when executed by the processor further performs the steps of:

In one embodiment, a computer program product is provided comprising a computer program which, when executed by a processor, performs the steps of:

It should be noted that, the user information (including, but not limited to, user equipment information, user personal information, etc.) and the data (including, but not limited to, data for analysis, stored data, presented data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the related data are required to comply with the related laws and regulations and standards of the related countries and regions.

Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the various embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magnetic random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase Change Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the various embodiments provided herein may include at least one of relational databases and non-relational databases. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic units, quantum computing-based data processing logic units, etc., without being limited thereto.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The foregoing examples represent only a few embodiments of the present application, which are described in more detail and are not thereby to be construed as limiting the scope of the present application. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application shall be subject to the appended claims.

Claims

1. A method of identifying quality, the method comprising:

constructing an original three-dimensional simulation environment corresponding to equipment to be detected according to the use scene of the equipment to be detected;

determining at least two groups of object attributes of the identified objects according to the original three-dimensional simulation environment; wherein the object pose attributes and/or object identity attributes of the different sets of object attributes are different;

According to the object attributes of each group, disposing the identified objects in the original three-dimensional simulation environment to obtain a target three-dimensional simulation environment;

2. The method according to claim 1, wherein the constructing an original three-dimensional simulation environment corresponding to the device to be detected according to a usage scenario of the device to be detected includes:

determining environmental impact factors of the equipment to be detected based on the use scene of the equipment to be detected; wherein the environmental impact factors include climate impact factors and/or light impact factors;

determining at least one set of test environmental parameters based on the environmental impact;

3. The method according to claim 1, wherein the method further comprises:

4. A method according to claim 3, wherein said determining the range of identification of the device to be detected in the original three-dimensional simulated environment further comprises:

5. The method of claim 1, wherein deploying the identified objects in the original three-dimensional simulated environment according to the respective sets of object properties comprises:

and deploying the identification object corresponding to each object attribute in the original three-dimensional simulation environment based on the deployment position.

6. The method according to claim 1, wherein the performing the recognition quality detection on the device to be detected to obtain a recognition quality detection result includes:

performing identification quality detection on the equipment to be detected, and determining an identification accuracy index and a response time index;

and determining the recognition quality detection result according to the recognition accuracy index and the response time index.

7. An identification quality detection apparatus, the apparatus comprising:

and the detection module is used for carrying out recognition quality detection on the equipment to be detected based on the target original three-dimensional simulation environment to obtain a recognition quality detection result.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 6 when the computer program is executed.

9. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 6.