CN112669387B

CN112669387B - Lamp holder position determining method, device, storage medium, program, and road side equipment

Info

Publication number: CN112669387B
Application number: CN202011589362.0A
Authority: CN
Inventors: 刘博�; 黄乐平
Original assignee: Apollo Zhilian Beijing Technology Co Ltd
Current assignee: Apollo Intelligent Connectivity Beijing Technology Co Ltd
Priority date: 2020-12-28
Filing date: 2020-12-28
Publication date: 2024-04-16
Anticipated expiration: 2040-12-28
Also published as: CN112669387A

Abstract

The application provides a lamp holder position determining method, a device, a storage medium, a program and road side equipment, which relate to intelligent traffic technology and unmanned technology and comprise the following steps: determining each circle in each image aiming at the acquired images of multiple frames including signal lamps, and determining the circle center of each circle; filtering each circle center to obtain an effective circle center; and determining the number of circle centers according to the number of the circle centers and determining the lamp cap position of the signal lamp in the image according to the number of the circle centers and the effective circle centers. In this kind of embodiment, can confirm the lamp holder position through discern circular mode in the image to avoid image acquisition device or signal lamp position to change, lead to can't obtain complete signal lamp based on preset mark information, and the scheme that this application provided only need discern circular in the image just can realize, this kind of mode consuming time is few, and required power is few.

Description

Lamp holder position determining method, device, storage medium, program, and road side equipment

Technical Field

The present disclosure relates to intelligent transportation technology and unmanned technology in image processing technology, and in particular, to a method, an apparatus, a storage medium, a program, and a roadside device for determining a lamp cap position.

Background

Signal lamps are seen everywhere in life, comprising a plurality of lamp heads in the signal lamp. Such as a traffic light comprising three lightheads.

In some application scenarios, state information of the signal lamp needs to be collected, such as collecting the color of the signal lamp. In one embodiment, an image including a signal may be acquired and the status of the signal determined from the acquired image.

When determining the state of the signal lamp based on the image, the position of the signal lamp in the image needs to be determined in advance, and then recognition is performed. However, in the actual process, no matter the thermal expansion and the cold contraction of the erection equipment or the ageing and loosening of the installation element can cause the image acquisition equipment to slightly move, and further, a complete signal lamp cannot be obtained according to a predetermined position, and an accurate signal lamp state cannot be obtained.

Disclosure of Invention

The application provides a lamp cap position determining method, device, storage medium, program and road side equipment, which are used for determining the lamp cap position of a signal lamp according to a shot image comprising the signal lamp.

According to a first aspect of the present application, there is provided a method for determining a position of a lamp cap, comprising:

determining each circle in each image aiming at the acquired images of multiple frames including signal lamps, and determining the circle center of each circle;

Filtering each circle center to obtain an effective circle center;

and determining the number of circle centers according to the number of the circle centers and determining the lamp cap position of the signal lamp in the image according to the number of the circle centers and the effective circle centers.

According to a second aspect of the present application, there is provided a lamp head position determining apparatus, comprising:

the circle center determining unit is used for determining each circle in each image aiming at the collected images of the multiple frames including the signal lamp and determining the circle center of each circle;

the filtering unit is used for filtering each circle center to obtain an effective circle center;

the quantity determining unit is used for determining the quantity of the circle centers according to the effective circle centers;

and the position determining unit is used for determining the lamp cap position of the signal lamp in the image according to the number of the circle centers and each effective circle center.

According to a third aspect of the present application, there is provided an electronic device comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of determining a lamp head position according to the first aspect.

According to a fourth aspect of the present application, there is provided a non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of determining a position of a lighthead according to the first aspect.

According to a fifth aspect of the present application, there is provided a computer program product comprising a computer program which, when executed by a processor, implements the method of determining a position of a burner as described in the first aspect.

According to a sixth aspect of the present application, there is provided a roadside apparatus comprising: an electronic device as shown in the third aspect.

The application provides a lamp holder position determining method, a device, a storage medium, a program and road side equipment, comprising the following steps: determining each circle in each image aiming at the acquired images of multiple frames including signal lamps, and determining the circle center of each circle; filtering each circle center to obtain an effective circle center; and determining the number of circle centers according to the number of the circle centers and determining the lamp cap position of the signal lamp in the image according to the number of the circle centers and the effective circle centers. In this kind of embodiment, can confirm the lamp holder position through discern circular mode in the image to avoid image acquisition device or signal lamp position to change, lead to can't obtain complete signal lamp based on preset mark information, and the scheme that this application provided only need discern circular in the image just can realize, this kind of mode consuming time is few, and required power is few.

It should be understood that the description of this section is not intended to identify key or critical features of the embodiments of the application or to delineate the scope of the application. Other features of the present application will become apparent from the description that follows.

Drawings

The drawings are for better understanding of the present solution and do not constitute a limitation of the present application. Wherein:

FIG. 1A is an application scenario diagram illustrating another exemplary embodiment of the present application;

FIG. 1B is a schematic view of an image acquired by an image acquisition device according to an exemplary embodiment of the present application;

FIG. 1C is a signal lamp according to an exemplary embodiment of the present application;

FIG. 1D is a signal lamp shown in another exemplary embodiment of the present application;

FIG. 2 is a flow chart of a method for determining a position of a light head according to an exemplary embodiment of the present application;

FIG. 3 is a flow chart illustrating a method of determining a position of a lighthead according to another exemplary embodiment of the present application;

FIG. 4A is a schematic diagram illustrating center filtering according to an exemplary embodiment of the present application;

FIG. 4B is a schematic diagram of clustering effective circle centers according to an exemplary embodiment of the present application;

FIG. 4C is a schematic diagram illustrating a determination of a class center according to an exemplary embodiment of the present application;

Fig. 5 is a schematic structural view of a lamp head position determining device according to an exemplary embodiment of the present application;

fig. 6 is a schematic structural view of a lamp head position determining device according to an exemplary embodiment of the present application;

fig. 7 is a block diagram of an electronic device for implementing a method of determining a lamp head position according to an embodiment of the present application.

Detailed Description

Exemplary embodiments of the present application are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present application to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present application. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

Signal lamps are arranged in many application scenes, and the guiding effect can be achieved through the signal lamps. In some embodiments, it is desirable to collect status information of the signal, such as the need to identify the light color of the signal, thereby facilitating the boot function. For example, in the intelligent traffic field, traffic such as red light running detection can be performed by identifying the color of the signal lamp, and the state of the current signal lamp can be broadcasted to vehicles at a distance, so that the vehicles can conduct path planning in advance.

However, when determining the state of the signal lamp based on the image, it is necessary to determine the position of the signal lamp in the image in advance and then to perform recognition. In the identification process, if the erection equipment is subjected to thermal expansion and contraction or the installation element is aged and loosened, the image acquisition equipment is moved, so that the complete signal lamp cannot be acquired in the acquired image based on the position calibrated in advance, and the accurate signal lamp state cannot be obtained.

The application provides a lamp cap position determining method which is applied to intelligent traffic technology and unmanned technology and is used for determining the lamp cap position of a signal lamp according to a shot image comprising the signal lamp.

In one embodiment, an image capture device may be positioned adjacent to the signal so that an image of the signal is captured using the image capture device, which may be a camera, through which successive images including the signal may be captured.

Further, the image acquisition device can send the shot image to a server for acquiring the signal lamp state, and the electronic equipment can process according to the received image based on the requirement.

Fig. 1A is an application scenario diagram illustrating another exemplary embodiment of the present application.

As shown in fig. 1A, a road side device 11 may be disposed beside a road, the road side device 11 is connected to an image acquisition device 12, and the image acquisition device 12 may capture an image of a signal lamp 13.

Fig. 1B is a schematic view of an image acquired by an image acquisition device according to an exemplary embodiment of the present application.

The image capturing apparatus 12 may capture an image as shown at 14 in fig. 1B and transmit the image 14 to the roadside device 11, or a server.

The area where the signal lamp is located may be labeled in advance, and the road side device 11 or the server may clip the received image 14 according to the signal lamp position labeled in advance, obtain an image including only the signal lamp as shown in fig. 15, and process the clipped image based on the requirement.

FIG. 1C is a signal lamp according to an exemplary embodiment of the present application; fig. 1D is a signal lamp shown in another exemplary embodiment of the present application.

As shown in fig. 1C and 1D, the signal lamp has a circular base frame in any of the modes. When the signal lamp image is processed, the lamp cap state in the signal lamp is mainly identified, so that the method provided by the application can avoid the problem that the signal lamp state cannot be obtained due to deviation of the marking position of the signal lamp by identifying the circle in the continuous multi-frame image and determining the lamp cap position in the image based on the identification result.

Fig. 2 is a flowchart illustrating a method for determining a position of a lamp head according to an exemplary embodiment of the present application.

As shown in fig. 2, a method for determining a lamp cap position according to an exemplary embodiment of the present application includes:

step 201, determining circles in each image according to the acquired images of multiple frames including signal lamps, and determining the circle center of each circle.

The method provided by the application can be performed by an electronic device with computing capability, and the electronic device can receive the image comprising the signal lamp acquired by the image acquisition device.

The electronic device may also process the received image based on requirements, for example, may perform a scheme of identifying the status of the signal lights, or extracting the color of the signal lights.

Specifically, when the electronic device processes the received image, the signal lamp included in the image can be extracted based on the pre-calibrated position information, so as to identify the signal lamp. If the position of the image acquisition device is deviated, or the equipment provided with the signal lamp is deviated, the complete signal lamp cannot be extracted based on the pre-calibrated position information.

Further, the electronic device may also determine the lighthead position in the image from the acquired images when processing based on the images.

In practical application, the electronic device may process the collected images including the signal lamp for multiple frames, for example, the number n of images may be set, and the collected continuous n frames of images may be processed.

Each time the electronic device receives an image, each circle in the image can be determined, and the center of each circle can be determined. The electronic device may further store the determined circle center in a preset list, for example, a circle center list may be preset, and the circle center list may store a circle center corresponding to each frame of image.

Specifically, the circle center may be recorded in a coordinate manner, for example, the image identifier P1 is recorded in the list, and circle centers (x 1, y 1), (x 2, y 2), and (x 3, y 3) corresponding to the image identifier P1 may also be recorded. Where x1, y1, x2, y2, x3, y3 may be pixel coordinate values in the image.

Further, the electronic device may determine a circle in the image using a hough transform algorithm.

The hough transform is a feature detection (feature extraction) algorithm, and is used to identify features in found objects, such as: circular.

Because the lamp cap of the signal lamp is circular, a circular position conforming to the lamp cap style can be determined in one frame of image by utilizing a mode of identifying the circular shape through Hough transformation.

In practical application, if the number of frames of the processed image reaches the preset number of frames, step 202 may be executed, otherwise, step 201 is executed on the received image frame continuously. For example, the preset number of frames is n, and step 202 may be performed after the circle centers of the n frames are obtained.

Step 202, filtering each circle center to obtain an effective circle center.

The circle centers identified in each image can be filtered to obtain effective circle centers. For example, if a center other than the lamp cap exists in the image, other centers than the lamp cap center may be obtained. In addition, due to the accuracy of the recognition algorithm or errors in the process of recognizing the images by the electronic equipment, the circle center recognition results of the same lamp cap in different images may be different. If the error of the center recognition result is larger in a certain image, the recognition result deviating from the center position of the lamp cap may appear.

Therefore, in the method provided by the application, the circle centers obtained through identification can be filtered. The lamp heads of the signal lamps are generally arranged on the same straight line, so that the circle centers of the lamp heads are positioned on the same straight line. Thus, it can be determined whether the identified circle centers are on the same straight line, and if most of the circle centers are on or near the straight line, the circle centers that are not near the straight line will be filtered.

In one embodiment, the obtained circle center information may be filtered based on a random sampling coincidence algorithm to obtain the effective circle center. The random sampling consistency algorithm can filter out the circle centers which do not meet the requirements, and then the effective circle center is obtained.

And 203, determining the number of circle centers according to the effective circle centers, and determining the lamp cap positions of the signal lamps in the image according to the number of circle centers and the effective circle centers.

The circle center number refers to the number of circle centers determined by the electronic device according to the multi-frame images, for example, if the number of circle centers is 3, the electronic device can be considered to determine that three circles exist in the images according to the multi-frame images, and if the number of circles is 2, the electronic device can be considered to determine that two circles exist in the images according to the multi-frame images.

Specifically, when the electronic device identifies the same circle in different images, the identified positions may also have differences, and thus the obtained circle centers may also have differences.

Further, the position recognition results of the same circle in different images should be similar even if there is a difference. For example, for a circle corresponding to the base 1, the corresponding center C1 may be identified in the first image, and the corresponding center C2 may be identified in the second image. C1 and C2 should be closer together. For another example, for a circle corresponding to the burner 2, the corresponding center C3 may be identified in the first image, C3 should be further away from C1.

Thus, the number of circle centers can be determined from the respective effective circle centers. For example, the electronic device may cluster the effective circle centers and determine the number of circle centers according to the clustering result. For example, if each effective center can be classified into 3 types, the electronic device determines the number of centers to be 3.

The electronic equipment can determine the lamp cap positions of the signal lamps in each image according to the number of the circle centers and the identified effective circle centers.

For example, if the number of determined circle centers is identical to the number of lamp bases of the signal lamp, the lamp base positions can be determined directly according to each effective circle center. If the number of the determined circle centers is not consistent with the number of the lamp caps, the determined effective circle centers can be deleted, and then the circle centers are continuously determined according to the newly received image, so that the lamp cap positions of the signal lamps in the image are determined again based on the newly determined circle centers and the existing effective circle centers.

Specifically, if the number of the determined circle centers is consistent with the number of the lamp caps of the signal lamps, the lamp cap positions of the signal lamps in the image can be determined according to the effective circle centers. For example, if the effective circle centers are clustered to obtain three classification results, the average positions of the effective circle centers in each class can be calculated to obtain three lamp cap positions.

The application provides a lamp cap position determining method, which comprises the following steps: determining each circle in each image aiming at the acquired images of multiple frames including signal lamps, and determining the circle center of each circle; filtering each circle center to obtain an effective circle center; and determining the number of circle centers according to the number of the circle centers and determining the lamp cap position of the signal lamp in the image according to the number of the circle centers and the effective circle centers. In this kind of embodiment, can confirm the lamp holder position through discern circular mode in the image to avoid image acquisition device or signal lamp position to change, lead to can't obtain complete signal lamp based on preset mark information, and the scheme that this application provided only need discern circular in the image just can realize, this kind of mode consuming time is few, and required power is few.

Fig. 3 is a flowchart illustrating a method for determining a position of a lamp head according to another exemplary embodiment of the present application.

As shown in fig. 3, the method for determining the position of the lamp cap provided by the present application includes:

step 301, determining circles in each image based on hough transform for the acquired multi-frame image including signal lamp, and determining the center of each circle.

Specifically, the electronic device may determine a circle in the image using a hough transform algorithm. The center of the circle can also be determined from the determined circle.

After the electronic device recognizes the circle in the image, the center of the circle can also be determined.

Step 302, filtering the obtained center information based on a random sampling coincidence algorithm to obtain an effective center.

After determining the circle centers corresponding to the multi-frame images, the electronic device can filter the obtained circle center information based on a random sampling consensus algorithm (RANdom SAmple Consensus, RANSAC).

Specifically, RANSAC may estimate parameters of the mathematical model in an iterative manner from a set of observations that contain "outliers". An outlier refers to a point that does not fit the model.

Because the lamp cap center of the signal lamp is positioned on a straight line, the lamp cap center point can adapt to a straight line model. The determined circle centers can be filtered based on RANSAC to find the effective circle centers conforming to the straight line model.

Further, RANSAC may select a set of subsets from the determined circle centers, and assume that the subset is an intra-office point. The current local point can be used for estimating a linear model, then the linear model is used for testing other circle centers except the local point, and if one circle center is suitable for the model, the circle center is considered to be the local point. If there are enough centers classified as hypothetical in-office points, then the estimated model can be considered accurate, and these centers classified as in-office points can be determined as valid centers.

In practical application, because the centers of all lamp caps in the signal lamp are positioned on the same straight line, the centers of all the lamp caps can be filtered through the RANSAC algorithm, and the obtained effective center accords with the position of the center of the lamp cap, so that when the position of the center of the lamp cap is determined based on the position of the center of the lamp cap, the determined result is more accurate.

Fig. 4A is a schematic diagram illustrating center filtering according to an exemplary embodiment of the present application.

As shown in fig. 4A, a plurality of circle centers can be identified from the multi-frame image, and each circle center 41 can be filtered according to its position.

As shown in fig. 4A, most of the circle centers can be determined to be adapted to the straight line model 41 based on the RANSAC algorithm, and the circle center 42 is not adapted to the model, so that the circle center 42 can be filtered out, and other circle centers are effective circle centers.

And 303, clustering each effective circle center according to the position of each effective circle center, and determining the number of categories in the clustering result as the number of circle centers.

After the effective circle centers are obtained, the electronic equipment can cluster the effective circle centers according to the positions of the effective circle centers.

Specifically, if a plurality of lightheads are included in an image, the effective center of a circle corresponding to each lighthead can be identified in each image. And the effective circle centers corresponding to the same lamp cap have aggregation tendency, so that the electronic equipment can aggregate the effective circle centers of a small-range area into one type.

If the effective circle centers determined in the multi-frame image are gathered into one type, it can be considered that the effective circle centers actually belong to the same circle center in the image, for example, the center of the same lamp cap is possible, but the circle centers are not completely overlapped due to algorithm or calculation errors. Therefore, the number of categories in the clustering result can be determined as the number of circle centers. That is, the effective circle centers in one category belong to the same circle center of the circle.

Fig. 4B is a schematic diagram of clustering effective circle centers according to an exemplary embodiment of the present application.

As shown in FIG. 4B, the points are effective circle centers determined by the electronic device, and the effective circle centers can be clustered, for example, categories 43, 44 and 45 can be obtained, and each category includes a plurality of effective circle centers.

The number of the actual circle centers in the image can be determined in this way, and then the number of the lamp caps can be determined.

And 304, if the number of the circle centers is 3 or 2, determining the lamp cap position of the signal lamp in the image according to the effective circle centers included in each category in the clustering result.

If the number of the determined circle centers is 3, it is indicated that the electronic device recognizes three circles in the image, that is, the center positions of the three lamp caps may be recognized. Under the condition, the electronic equipment can determine the lamp cap position of the signal lamp in the image according to the effective circle centers included in each category.

If the number of the determined circle centers is 2, the electronic equipment identifies two circles in the image, namely the central positions of the two lamp caps can be identified. Under the condition, the electronic equipment can determine the lamp cap position of the signal lamp in the image according to the effective circle centers included in each category.

Specifically, if the number of circle centers is 3 or 2, the electronic device can determine whether the circle centers conform to the position of the lamp cap of the signal lamp. If the number of the determined circle centers is 1, because only one circle center exists, whether the circle center accords with the position of the lamp cap of the signal lamp or not cannot be determined, and therefore, when the number of the circle centers is 1, the scheme provided by the application does not determine the position of the lamp cap in the image based on the currently determined effective circle centers, and the determination result of the lamp cap is more accurate.

Further, when determining the lamp cap positions of the signal lamps in the image according to the effective circle centers included in each category in the clustering result, the electronic device can combine the size information of the signal lamps to determine whether the effective circle centers accord with the positions of the lamp caps of the signal lamps.

In practical application, the electronic device can acquire the size information of the signal lamp, and can also determine the class center corresponding to each class according to the effective circle centers included in each class in the clustering result.

The size information can be marked in advance, and the electronic equipment can directly acquire the size information of the signal lamp when determining the position of the lamp cap. For example, an image including the signal lamp can be acquired by the image acquisition device, and then the position of the signal lamp is marked by the user, so that the electronic equipment can acquire the marking information, and the size information of the signal lamp is determined according to the marking information. In this way, the size information of the signal lamp can be determined according to the signal lamp frame marked in advance, and whether the identified category center accords with the lamp cap position of the signal lamp can be further determined according to the size information of the signal lamp.

Specifically, the electronic device may process the effective circle center included in each category in the clustering result, to obtain a category center corresponding to each category. For example, the average position of each effective center in a category may be determined to obtain the category center for that category.

Fig. 4C is a schematic diagram illustrating a determination manner of a category center according to an exemplary embodiment of the present application.

As shown in fig. 4C, the clustering result includes three categories 46, 47, 48, each including a number of effective centers. A category center corresponding to each category may be determined based on the effective center of the circle in that category.

For example, category 46 corresponds to category center 461, category 47 corresponds to category center 471, and category 48 corresponds to category center 481.

Further, the electronic device can determine the lamp cap position of the signal lamp in the image according to the acquired size information and the determined category center. Specifically, whether the category center accords with the size information of the signal lamp can be determined, and if so, the lamp cap position of the signal lamp can be determined according to the determined category center.

In the embodiment, whether the currently identified category center is accurate or not can be determined by combining the actual size information of the signal lamp, so that the fact that other circular interference identification results except the lamp cap exist in the image is avoided.

In one embodiment, if the number of circle centers is 2, the center-to-center distance between two category centers can be determined according to the category centers; and according to the size information and the center-to-center distance, and determining the lamp cap position of the signal lamp in the image.

If the number of the determined circle centers is 2, two clustering categories are indicated, and if the category centers corresponding to the two clustering categories are in accordance with the lamp cap positions of the signal lamps, the lamp cap positions of the signal lamps in the image can be determined according to the category centers corresponding to the clustering categories.

Specifically, the center-to-center distance between two category centers can be determined, and whether the category centers accord with the lamp cap positions of the signal lamps in the image can be determined according to the center-to-center distance.

In the embodiment, whether the identification result accords with the lamp cap position of the signal lamp can be determined according to the center-to-center distance of the category center, the method is simple and easy to implement, and the lamp cap position of the signal lamp can be determined without consuming much calculation force.

Specifically, if the difference between the center distance and two thirds of the height value of the signal lamp is not more than a threshold value, or the difference between the center distance and two thirds of the width value of the signal lamp is not more than a threshold value, determining the lamp cap position of the signal lamp in the image according to the category center.

Further, if the difference between the center distance of the two category centers and two thirds of the height value of the signal lamp does not exceed a threshold value, the two category centers are two lamp cap centers at two ends of three lamp caps which are vertically arranged; if the center distance between the centers of the two categories and the two thirds of the width value of the signal lamp are not more than a threshold value, the two category centers are two lamp cap centers at two ends of the three lamp caps which are transversely arranged.

In practical application, the lamp caps in the signal lamp are uniformly distributed, so that if the positions of the two lamp caps at the two ends of the signal lamp are determined, the middle lamp cap position can be determined. In this embodiment, after identifying the two category centers, it may be determined whether the two category centers conform to the positions of the bases at the two ends of the signal lamp, so that when the conditions may be met, the electronic device may determine the positions of the bases in the signal lamp.

If the center-to-center distance between the two category centers meets the above condition, when determining the lamp cap positions according to the two category centers, the first category center can be determined as the center position of the first lamp cap of the signal lamp in the image, and the second category center can be determined as the center position of the third lamp cap of the signal lamp in the image.

Specifically, if the center-to-center distance between the two category centers meets the above condition, it can be stated that the determined category center is the center of the two bases at both ends in the signal lamp, and therefore, the two category centers can be directly determined as the center positions of the bases at both ends of the signal lamp.

Further, the position of the middle lamp cap of the signal lamp can be determined according to the determined two category centers. Specifically, the middle position can be determined according to the positions of the centers of the two categories, and the middle position is determined as the center position of the middle lamp cap of the signal lamp.

In practical application, the embodiment can determine the lamp cap position of the signal lamp under the condition that two circle center positions are determined, so that a complete signal lamp image can be obtained in a shot image under the condition that the relative positions of the image acquisition device and the signal lamp are offset.

In another embodiment, if the number of circle centers is 3, the center-to-center distance between every two category centers can be determined according to the category centers; and determining the lamp cap position of the signal lamp in the image according to the size information and the center distance.

If the number of the determined circle centers is 3, it is indicated that three clustering categories exist, and if the category centers corresponding to the three clustering categories are in accordance with the lamp cap positions of the signal lamps, the lamp cap positions of the signal lamps in the image can be determined according to the category centers corresponding to the clustering categories.

Specifically, the center-to-center distance between every two category centers can be determined, and whether the category centers accord with the lamp cap positions of the signal lamps in the image can be determined according to the center-to-center distance.

Specifically, if any determined center-to-center distance meets any of the following conditions, determining the category center as the center position of the lamp cap center of the signal lamp in the image:

the difference between the center distance and two thirds of the height value of the signal lamp is not more than a threshold value;

the difference between the center distance and one third of the height value of the signal lamp is not more than a threshold value;

the difference between the center distance and two thirds of the signal lamp width value does not exceed a threshold value;

the center-to-center spacing does not differ from one third of the signal lamp width by more than a threshold value.

Further, if any center-to-center distance between the centers of the three categories meets any of the above conditions, it is indicated that the center-to-center distance is the distance between two bases of the signal bases that are closer to each other, or the center-to-center distance is the distance between two bases of the signal bases that are farther to each other. Thus, it can be determined that the identified category center coincides with the cap position of the signal lamp.

In this embodiment, after three category centers are identified, it may be determined whether the three category centers match the positions of the bases of the signal lamps, so that when the conditions match, the electronic device may be able to determine the base positions in the signal lamps.

If the center-to-center distances of the determined three category centers meet the above conditions, the three category centers can be directly determined as the center positions of the three lamp caps.

Step 305, if the number of circle centers is not 2 or 3, deleting the circle center with the earliest time according to the determined sequence of the circle centers, and collecting a new image comprising the signal lamp; each circle in the new image is determined and the center of the circle is determined.

After step 305, step 302 may continue to be performed.

Further, if the number of circle centers is not 2 or 3, the electronic device may delete part of the circle centers. And re-determining the circle center according to the newly acquired image.

In practical application, if the number of circle centers is not 2 or 3, the currently identified circle centers can be considered to have centers except the lamp cap of the signal lamp, so that part of the circle centers can be deleted. For example, the center of the circle corresponding to the earliest received image may be deleted.

In this embodiment, the lamp cap position can be redetermined by combining the existing data and the newly acquired image under the condition that the lamp cap position identification fails, so that the target circle center position which accords with the lamp cap position of the signal lamp can be found.

Step 306, the initial center position is obtained.

In an alternative embodiment, the electronic device may also obtain the initial center position. The initial center position refers to a pre-labeled center position of the lamp cap, for example, the center position may be determined at the beginning of the method provided in this embodiment. For example, the electronic device may determine the initial center position based on a pre-noted signal light bezel position.

Step 307, determining the offset information of the lamp cap position according to the determined lamp cap position and the initial circle center position.

After determining the lamp cap position, the electronic device can compare the determined lamp cap position with the initial circle center position, so that the offset information of the lamp cap position is determined.

Specifically, the electronic device may calculate the difference between the base position and the initial center position according to the pixel coordinate values of the two positions. If a plurality of lamp caps exist, the difference between the corresponding lamp cap positions and the initial circle center positions can be calculated, and a plurality of position differences are obtained. The electronic device may determine an average value between the plurality of position differences, and determine the offset information for the lamp head position.

In this embodiment, the electronic device may determine the offset information of the lamp cap, and further may obtain the complete signal lamp in the image according to the offset information when processing the acquired image based on the pre-labeled signal lamp frame.

In any of the above embodiments, the signal lamp frame positions may be marked in the history image in advance, and the electronic device may determine the initial center position according to the signal lamp frame positions and the preset number of lamp caps.

The central position of each lamp cap, namely the initial circle center position, can be determined according to the positions of the lamp frames of the signal lamps and the number of preset lamp caps. For example, when the number of the lamp bases marked in the history image is 3, the initial center positions (light_1_center_x, light_1_center_y), (light_2_center_x, light_2_center_y), and (light_3_center_x, light_3_center_y) corresponding to the three lamp bases can be determined.

In the embodiment, the electronic equipment does not need to store the coordinate positions of the lamp caps, and only needs to determine the initial circle center position by utilizing the pre-marked signal lamp frame positions when acquiring the initial circle center position, so that the storage space can be saved.

Fig. 5 is a schematic structural view of a lamp head position determining device according to an exemplary embodiment of the present application.

As shown in fig. 5, the determining device 500 for determining a lamp cap position provided in the present application includes:

a circle center determining unit 510, configured to determine, for a plurality of collected images including signal lamps, circles in each of the images, and determine a circle center of each of the circles;

The filtering unit 520 is configured to filter each circle center to obtain an effective circle center;

a number determining unit 530, configured to determine the number of circle centers according to each effective circle center;

and the position determining unit 540 is configured to determine the lamp cap position of the signal lamp in the image according to the number of circle centers and each effective circle center.

The lamp cap position determining device comprises a circle center determining unit, a circle center determining unit and a lamp cap position determining unit, wherein the circle center determining unit is used for determining each circle in each image aiming at the collected images of multiple frames including signal lamps and determining the circle center of each circle; the filtering unit is used for filtering each circle center to obtain an effective circle center; the quantity determining unit is used for determining the quantity of the circle centers according to the effective circle centers; and the position determining unit is used for determining the lamp cap position of the signal lamp in the image according to the number of the circle centers and each effective circle center. In this kind of embodiment, can confirm the lamp holder position through discern circular mode in the image to avoid image acquisition device or signal lamp position to change, lead to can't obtain complete signal lamp based on preset mark information, and the scheme that this application provided only need discern circular in the image just can realize, this kind of mode consuming time is few, and required power is few.

Fig. 6 is a schematic structural view of a lamp head position determining device according to an exemplary embodiment of the present application.

As shown in fig. 6, the determining device 600 for a lamp cap position provided in the present application, wherein the filtering unit 620 is specifically configured to:

and filtering the obtained circle center information based on a random sampling consistency algorithm to obtain the effective circle center.

Wherein the number determining unit 630 includes:

the clustering module 631 is configured to cluster each effective circle center according to the position of each effective circle center;

the number determining module 632 is configured to determine the number of categories in the clustering result as the number of circle centers.

The location determining unit 64 specifically is configured to:

and if the number of the circle centers is 3 or 2, determining the lamp cap position of the signal lamp in the image according to the effective circle centers included in each category in the clustering result.

Wherein the position determining unit 640 includes:

an acquisition module 641 for acquiring size information of the signal lamp;

a center determining module 642, configured to determine a category center corresponding to each category according to the effective circle centers included in each category in the clustering result;

and the position determining module 643 is configured to determine a lamp cap position of the signal lamp in the image according to the size information and the category center.

If the number of circle centers is 2, the location determining module 643 is specifically configured to:

determining the center-to-center distance between two category centers according to the category centers;

and determining the lamp cap position of the signal lamp in the image according to the size information and the center distance.

The location determining module 643 is specifically configured to:

and if the difference between the center distance and the two thirds of the height value of the signal lamp is not more than a threshold value, or the difference between the center distance and the two thirds of the width value of the signal lamp is not more than a threshold value, determining the lamp cap position of the signal lamp in the image according to the category center.

The location determining module 643 is specifically configured to:

determining a first type center as the center position of a first lamp cap of the signal lamp in the image, and determining a second type center as the center position of a third lamp cap of the signal lamp in the image;

and determining the center position of a second lamp cap of the signal lamp in the image according to the first class center and the second class center.

If the number of circle centers is 3, the location determining module 643 is specifically configured to:

determining the center distance between the class centers according to the class centers;

The location determining module 643 is specifically configured to:

if any of the determined center-to-center distances meets any of the following conditions, determining the category center as the center position of the lamp cap of the signal lamp in the image:

the center-to-center distance is different from two thirds of the height value of the signal lamp by no more than a threshold value;

the center-to-center distance is different from one third of the height value of the signal lamp by no more than a threshold value;

the difference between the center distance and two thirds of the width value of the signal lamp is not more than a threshold value;

the center-to-center spacing does not differ from one third of the signal lamp width value by more than a threshold value.

The method comprises the steps that a signal lamp frame position is marked in a historical image in advance;

the obtaining module 641 is specifically configured to:

and determining the size information of the signal lamp according to the position of the signal lamp frame.

Wherein if the number of circle centers is not 2 or 3, the apparatus further comprises:

a deleting unit 650, configured to delete the circle center with the earliest determined time according to the determined order of the circle centers, and collect a new image including the signal lamp;

the circle center determining unit 610 is further configured to determine each circle in the new image, and determine a circle center of the circle;

The filtering unit 620 is further configured to perform the step of filtering each circle center with respect to the existing circle center to obtain an effective circle center.

Wherein further comprising an offset determination unit 660 for:

acquiring an initial circle center position;

and determining the offset information of the lamp cap position according to the determined lamp cap position and the initial circle center position.

the acquisition module 641 and/or the offset determination unit 660 are specifically configured to:

and determining the initial circle center position according to the signal lamp frame positions and the preset lamp cap number.

The center determining unit 610 is specifically configured to:

circles are determined in each of the images based on a hough transform.

In an alternative embodiment, the instant application further provides a computer program product comprising a computer program stored in a readable storage medium, from which at least one processor of an electronic device can read the computer program, the at least one processor executing the computer program causing the electronic device to perform the scheme provided by any of the embodiments described above.

The application also provides a roadside device, comprising: an electronic device is shown below.

According to embodiments of the present application, an electronic device and a readable storage medium are also provided.

Fig. 7 shows a schematic block diagram of an example electronic device 800 that may be used to implement embodiments of the present application. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 7, the electronic device 700 includes a computing unit 701 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 702 or a computer program loaded from a storage unit 708 into a Random Access Memory (RAM) 703. In the RAM703, various programs and data required for the operation of the device 700 may also be stored. The computing unit 701, the ROM702, and the RAM703 are connected to each other through a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.

Various components in device 700 are connected to I/O interface 705, including: an input unit 706 such as a keyboard, a mouse, etc.; an output unit 707 such as various types of displays, speakers, and the like; a storage unit 708 such as a magnetic disk, an optical disk, or the like; and a communication unit 709 such as a network card, modem, wireless communication transceiver, etc. The communication unit 709 allows the device 700 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.

The computing unit 701 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 701 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The calculation unit 701 performs the respective methods and processes described above, for example, a determination method of the cap position. For example, in some embodiments, the method of determining the position of the lighthead may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as the storage unit 708. In some embodiments, part or all of the computer program may be loaded and/or installed onto device 700 via ROM 702 and/or communication unit 709. When a computer program is loaded into the RAM 703 and executed by the computing unit 701, one or more steps of the above-described method of determining a cap position may be performed. Alternatively, in other embodiments, the computing unit 701 may be configured to perform the method of determining the burner position by any other suitable means (e.g. by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.

The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service ("Virtual Private Server" or simply "VPS") are overcome. The server may also be a server of a distributed system or a server that incorporates a blockchain.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present application may be performed in parallel, sequentially, or in a different order, provided that the desired results of the technical solutions disclosed in the present application can be achieved, and are not limited herein.

The above embodiments do not limit the scope of the application. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application are intended to be included within the scope of the present application.

Claims

1. A method of determining a position of a lamp head, comprising:

filtering each circle center to obtain an effective circle center;

determining the number of circle centers according to the effective circle centers, and determining the lamp cap position of the signal lamp in the image according to the number of the circle centers and the effective circle centers;

The determining the number of circle centers according to each effective circle center comprises the following steps:

clustering each effective circle center according to the position of each effective circle center;

determining the number of categories in the clustering result as the number of circle centers;

the determining the lamp cap position of the signal lamp in the image according to the number of the circle centers and each effective circle center comprises the following steps:

if the number of the circle centers is 3 or 2, determining the lamp cap position of the signal lamp in the image according to the effective circle centers included in each category in the clustering result;

determining the lamp cap position of the signal lamp in the image according to the effective circle centers included in each category in the clustering result comprises the following steps:

acquiring size information of a signal lamp;

determining a class center corresponding to each class according to the effective circle centers included in each class in the clustering result;

and determining the lamp cap position of the signal lamp in the image according to the size information and the class center.

2. The method of claim 1, wherein filtering each center of a circle to obtain an effective center of a circle comprises:

3. The method of claim 1, wherein if the number of circle centers is 2, determining a lamp cap position of a signal lamp in the image according to the size information and the category center comprises:

4. A method according to claim 3, wherein determining the cap position of the signal lamp in the image from the size information, the center-to-center distance, comprises:

5. The method of claim 4, wherein determining a cap position of a signal lamp in the image from the category center comprises:

6. The method of claim 1, wherein if the number of circle centers is 3, determining a lamp cap position of a signal lamp in the image according to the size information and the category center comprises:

7. The method of claim 6, wherein determining a cap position of a signal lamp in the image based on the size information, each of the center-to-center distances, comprises:

8. The method of claim 1, wherein the signal lamp bezel locations are previously noted in the historical image;

acquiring size information of a signal lamp, comprising:

9. The method of any of claims 1-8, wherein if the number of centers is not 2 or 3, the method further comprises:

deleting the circle center with earliest time according to the circle center determining sequence, and collecting new images comprising signal lamps;

determining each circle in the new image, and determining the center of the circle;

and executing the step of filtering each circle center to obtain an effective circle center aiming at the existing circle centers.

10. The method of any of claims 1-8, further comprising:

acquiring an initial circle center position;

11. The method of claim 10, wherein the signal lamp bezel locations are previously noted in the historical image;

the obtaining the initial circle center position comprises the following steps:

12. The method of any of claims 1-8, wherein said determining the respective circles in each of said images comprises:

circles are determined in each of the images based on a hough transform.

13. A lamp head position determining apparatus comprising:

the position determining unit is used for determining the lamp cap position of the signal lamp in the image according to the number of the circle centers and each effective circle center;

the number determination unit includes:

the clustering module is used for clustering each effective circle center according to the position of each effective circle center;

the quantity determining module is used for determining the quantity of categories in the clustering result as the circle center quantity;

the position determining unit is specifically configured to:

the position determining unit further includes:

the acquisition module is used for acquiring the size information of the signal lamp;

the center determining module is used for determining a category center corresponding to each category according to the effective circle centers included in each category in the clustering result;

And the position determining module is used for determining the lamp cap position of the signal lamp in the image according to the size information and the category center.

14. The device according to claim 13, wherein the filtering unit is specifically configured to:

15. The apparatus of claim 14, wherein if the number of circle centers is 2, the position determining module is specifically configured to:

16. The apparatus of claim 15, wherein the location determination module is specifically configured to:

17. The apparatus of claim 16, wherein the location determination module is specifically configured to:

18. The apparatus of claim 13, wherein if the number of centers is 3, the position determining module is specifically configured to:

19. The apparatus of claim 18, wherein the location determination module is specifically configured to:

20. The apparatus of claim 13, wherein the signal lamp bezel locations are previously noted in the historical images;

the acquisition module is specifically used for:

21. The apparatus of any of claims 13-20, wherein if the number of centers is not 2 or 3, the apparatus further comprises:

the deleting unit is used for deleting the circle center with earliest determined time according to the determined sequence of the circle centers and collecting new images comprising the signal lamps;

the circle center determining unit is also used for determining each circle in the new image and determining the circle center of the circle;

the filtering unit is also used for executing the step of filtering each circle center to obtain an effective circle center aiming at the existing circle center.

22. The apparatus according to any one of claims 13-20, further comprising an offset determination unit configured to:

acquiring an initial circle center position;

23. The apparatus of claim 22, wherein the signal lamp bezel locations are previously noted in the historical images;

The acquisition module and/or the offset determination unit are specifically configured to:

24. The device according to any one of claims 13-20, wherein the centre determining unit is specifically configured to:

circles are determined in each of the images based on a hough transform.

25. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-12.

26. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-12.

27. A computer program product comprising a computer program which, when executed by a processor, implements the method of any of claims 1-12.

28. A roadside apparatus comprising: an electronic device as set forth in claim 25.