Photographing method, apparatus, system, and electronic device

Abstract

Embodiments of the present application provide a photographing method, device, system and electronic equipment, wherein the method comprises: determining a to-be-photographed person set and person information based on a first image photographed by a first camera; determining a photographable person set of the second camera at each rotation angle from the to-be-photographed person set based on camera parameters of the second camera and position information in the person information, the first camera and the second camera being arranged at different positions; determining a target angle from the rotation angles based on the photographable person set at each rotation angle, and controlling the second camera to photograph at the target angle. The present application determines the photographable person set and the person information of the low-point camera at different rotation angles, determines the optimal rotation angle, and controls the low-point camera to photograph at the optimal rotation angle, thereby avoiding the missed photographing of the face photos of the persons caused by too many blind areas or unsuitable camera angles, and improving the overall face photographing rate of the crowd.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, and in particular to a photographing method, device, system and electronic device. BACKGROUND

[0002] After many cameras are installed and the corresponding angles are adjusted, the photographing angles of the cameras are not adjusted in real time, so that when multiple personnel pass through the camera from different angles in some continuous road sections, the face information of the personnel cannot be captured, which leads to inaccurate data obtained when searching for personnel or drawing the face trajectory of a person, resulting in data deviation.

[0003] In related technologies, the rotation angle of a face in a current face picture compared to a front face is determined by pictures of faces at different angles captured by multiple cameras, and the smallest rotation angle is taken as the most front face picture. In fact, the final obtained front face picture is not necessarily a completely front face picture, and the face picture used in the process is not necessarily the picture of the person himself, and the front face capture rate is still not high. SUMMARY

[0004] To solve the above problems in the related art, embodiments of the present application provide a photographing method, device, system and electronic device.

[0005] In a first aspect, the present application provides a photographing method, comprising:

[0006] Based on a first image captured by a first camera, a set of to-be-photographed personnel and personnel information of each personnel in the set of to-be-photographed personnel are determined;

[0007] Based on the camera parameters of a second camera and the position information in the personnel information, a set of photographable personnel at each rotation angle of the second camera is determined from the set of to-be-photographed personnel, and the first camera and the second camera are arranged at different positions;

[0008] Based on the set of photographable personnel at each rotation angle, a target angle is determined from the rotation angles, and the second camera is controlled to photograph at the target angle.

[0009] In some embodiments, the target angle is determined from the rotation angles based on the set of photographable personnel at each rotation angle, comprising:

[0010] Based on at least one of the position, speed, face acquisition rate and face library comparison degree of compliance of each personnel in the set of to-be-photographed personnel, a photographing weight of the personnel is determined;

[0011] determine a shooting credit of each of the rotation angles based on the shooting weight of each of the persons in the set of persons who can be shot at each of the rotation angles;

[0012] determine the target angle from the rotation angles based on the shooting credit of each of the rotation angles.

[0013] In some embodiments, the determining the shooting weight of each of the persons in the set of persons to be shot based on at least one of the position, the speed, the face acquisition rate, and the face database comparison degree of each of the persons, comprises:

[0014] determine a distance between the person and a center point of a shooting area corresponding to the first camera based on the position of the person and the position of the center point;

[0015] determine the shooting weight of the person based on the distance and the number of the second cameras corresponding to the set of persons who can be shot.

[0016] In some embodiments, the determining the shooting weight of the person based on the distance and the number of the second cameras corresponding to the set of persons who can be shot, comprises:

[0017] determine a first weight of the person based on the distance and the number of the second cameras corresponding to the set of persons who can be shot;

[0018] determine a second weight of the person based on at least one of the speed, the face acquisition rate, and the face database comparison degree of the person;

[0019] determine the shooting weight of the person based on the first weight and the second weight of the person.

[0020] In some embodiments, the determining the face acquisition rate comprises:

[0021] determine a number of the second cameras that have historically shot the person as a person in a set of persons who can be shot and a number of times of shooting that have historically shot the person as a person in a set of persons who can be shot;

[0022] determine the face acquisition rate of the person based on the number of the second cameras and the number of times of shooting.

[0023] In some embodiments, the determining each of the rotation angles comprises:

[0024] determine each of the rotation angles in turn with the position of the person in the set of persons to be shot as a shooting center in a sequence from high to low of the shooting weight.

[0025] In some embodiments, the method further comprises:

[0026] obtaining a second image captured by the second camera at the target angle;

[0027] determining a photographed person from the photographable person set at the target angle based on the second image, and re-controlling the first camera to capture a new first image;

[0028] determining the photographable person set at each rotation angle based on the first image captured by the first camera, includes:

[0029] determining the photographable person set excluding the photographed person based on the new first image.

[0030] In a second aspect, the present application provides a photographing device, including:

[0031] a first determining unit, configured to determine a photographable person set and person information of each person in the photographable person set based on a first image captured by a first camera;

[0032] a second determining unit, configured to determine a photographable person set at each rotation angle of a second camera based on camera parameters of the second camera and position information in the person information, the first camera and the second camera being arranged at different positions;

[0033] a third determining unit, configured to determine a target angle from the rotation angles based on the photographable person set at each rotation angle, and control the second camera to capture an image at the target angle.

[0034] In a third aspect, the present application provides a photographing system, including a first camera, a second camera and the photographing device provided in the second aspect.

[0035] In a fourth aspect, the present application provides an electronic device, including a memory, a processor and a computer program stored in the memory and executable on the processor, the processor executes the program to implement the photographing method in any of the above aspects.

[0036] In a fifth aspect, the present application provides a non-transitory computer readable storage medium, having a computer program stored thereon, the computer program being executable on a processor to implement the photographing method in any of the above aspects.

[0037] In a sixth aspect, the present application provides a computer program product, including a computer program, the computer program being executable on a processor to implement the photographing method in any of the above aspects.

[0038] The photographing method, device, system and electronic equipment provided by the embodiments of the present application determine the optimal rotation angle of the low-point camera by determining the photographable person set of the low-point camera at different rotation angles, thereby controlling the low-point camera to photograph at the optimal rotation angle, avoiding the missed photographing of the face photos of the personnel caused by too many blind spots or unsuitable camera angles, and thereby improving the whole crowd face snapshot rate. BRIEF DESCRIPTION OF DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0040] Figure 1 is a flowchart of the photographing method provided by the embodiments of the present application;

[0041] Figure 2 is a schematic diagram of the space region photographable by the low-point camera provided by the embodiments of the present application;

[0042] Figure 3 is a structural schematic diagram of the photographing device provided by the embodiments of the present application;

[0043] Figure 4 is a structural schematic diagram of the photographing system provided by the embodiments of the present application;

[0044] Figure 5 is a structural schematic diagram of the electronic equipment provided by the embodiments of the present application. DETAILED DESCRIPTION

[0045] After the current many cameras are installed and the corresponding angles are adjusted, the photographing angles of the cameras will not be adjusted in real time, which results in that when multiple personnel pass through the camera from different angles in some continuous road sections, the face information of the personnel cannot be captured, which results in that the data obtained when searching for a person or drawing the face track of a person is not accurate, and the data deviation occurs.

[0046] In the related art, a most frontal face image output method and output system based on a multi-camera array are provided. The method comprises: acquiring multiple face images of a current face captured by the multi-camera array at a time; analyzing the rotation angle of the current face in each face image, and selecting a face image with the smallest rotation angle; determining whether the rotation angle of the current face in a newly acquired face image of the multi-camera array when tracking the current face is smaller than the selected face image with the smallest rotation angle; if so, replacing the selected face image with the smallest rotation angle with the newly acquired face image, and outputting a final face image as the most frontal face image of the current face when the current face leaves the tracking range.

[0047] The method selects the most frontal face based on space and time from the face images acquired by the multi-camera array, and the output most frontal face image is more accurate, thereby providing a more reliable data basis for subsequent face comparison.

[0048] However, the method determines the rotation angle of the face in the current face image compared with the frontal face from the face images captured by multiple cameras, and selects the smallest rotation angle as the most frontal face. In fact, the final acquired frontal face is not necessarily a complete frontal face image, and the face image used in the process is not necessarily the photo of the person, and the frontal face capture rate is still not high.

[0049] To solve the above problems in the related art, the embodiment of the present application provides a shooting method, device, system and electronic equipment. The optimal rotation angle of the low-point camera is determined by determining the set of people who can be shot by the low-point camera at different rotation angles, so that the low-point camera is controlled to shoot at the optimal rotation angle, and the missing shooting of the face photo of the person caused by too many blind spots or unsuitable camera angles is avoided, thereby improving the overall crowd face capture rate.

[0050] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely below with reference to the drawings in the present application. Obviously, the described embodiments are some of the embodiments of the present application, but not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0051] Figure 1 is a flowchart of the shooting method provided by the embodiment of the present application, as shown in Figure 1 The method comprises the following steps:

[0052] Step 101: determining a set of people to be shot and the personnel information of each person in the set of people to be shot based on the first image shot by the first camera.

[0053] Specifically, the first camera is a high-point camera. A high-point camera is a camera installed at a high altitude, such as the rooftop of a building, to provide a bird's-eye view of an area. A high-point camera array refers to multiple high-point cameras installed in different locations, which can capture images of the same area from multiple angles.

[0054] Images taken by high-point cameras are usually from a top-down perspective and can capture information about people, such as their location, height, clothing features, shoe features, whether they are carrying pets, and whether they are wearing hats.

[0055] Based on the first image captured by the first camera, a set of people to be photographed is determined, that is, the set of people to be photographed is determined from the area covered by the high-point camera array. Optionally, personnel information of each person in the set of people to be photographed can be obtained.

[0056] For example, at a certain time T, the personnel information obtained within the shooting area of ​​the high-point camera matrix is ​​shown in Table 1.

[0057] Table 1: Personnel Information Acquired by High-Point Cameras

[0058]

[0059]

[0060] The above personnel information only lists some common types of personnel information, and this application embodiment does not impose any restrictions on them.

[0061] Step 102: Based on the camera parameters of the second camera and the location information in the personnel information, determine the set of people that can be photographed by the second camera at various rotation angles from the set of people to be photographed. The first camera and the second camera are set in different positions.

[0062] Specifically, the first camera and the second camera are positioned differently. Optionally, the first camera is a high-point camera, and the corresponding second camera is a low-point camera. Furthermore, the low-point camera is a low-point camera within the coverage area of ​​the high-point camera array.

[0063] The camera parameters of the second camera, including but not limited to the installation position of the low-point camera, the rotation angle (horizontal rotation angle and pitch rotation angle), and the zoom factor, are shown in Table 2.

[0064] Table 2: Parameters of Low-Point Cameras

[0065]

[0066] By combining the camera parameters of the second camera with the position information of each person in the set of people to be photographed, the set of people that can be photographed by the second camera at various rotation angles is determined from the set of people to be photographed.

[0067] For any low point camera, the installation position is fixed, the rotatable angle and the zoomable multiple are determined, then the space region that can be shot at any rotatable angle is determined. In combination with the position information of the person, it can be determined whether the head of the person is in the space region that can be shot by the low point camera, so as to judge whether the person belongs to the shootable person of the low point camera. Alternatively, the position information of the person includes the latitude and longitude information where the person is located, and the height (height) information of the person.

[0068] Figure 2 is a schematic diagram of the space region that can be shot by the low point camera provided by the embodiment of the present application, as shown in Figure 2 , the geometric body ABCDE is the space region that can be shot by the low point camera at the rotatable angle perpendicular to the ground (trapezoid ABCD) and perpendicular to the wall surface (triangle ECD), and E is the installation position of the low point camera.

[0069] It can be seen that the camera is not a plane in actual shooting, but a three-dimensional geometric figure, that is, the geometric body ABCDE. To judge whether the head of the person is in the space region that can be shot by the low point camera, the following two conditions should be met:

[0070] (1) The latitude and longitude of the person need to be in the coverage range of the trapezoidal region ABCD.

[0071] (2) The height of the person cannot exceed the space region covered by the geometric body ABCDE. In combination with Figure 2 , for the person at the latitude and longitude position H (x h ,y h ), the highest point of the space region that can be shot by the low point camera is the G point, if the height of the person does not exceed the distance between H and G, it is considered that the head of the person is shootable, that is, the face of the person may exist in the shooting range of the low point camera. The GH height can be obtained by combining the prior art. The height of the person can be calculated by comparing with the surrounding reference in the first image shot by the high point camera.

[0072] When the low point camera is vertically or horizontally rotated, the space region that can be shot by the low point camera changes, and the GH height can also be obtained by the above-mentioned manner, so as to judge whether the person belongs to the shootable person of the low point camera.

[0073] Through the above calculation, at a moment Tn, according to the position where the person is located in the shooting region (which can be simply referred to as the high point region) of the high point camera array and the height information of the person, the shootable person set that can be shot by the head of the person by any low point camera at any shooting angle can be obtained.

[0074] For example, the total number of people in the area is 10, the low point camera A can capture 3 people at the current time, i.e., the horizontal angle A and the pitch angle A, and the personnel marks P1, P2 and P3. If the horizontal angle B and the pitch angle B can capture P1, P3 and P4. If the horizontal angle C and the pitch angle C capture P5 and P6. The number of people that can be captured by the three rotation angles and the personnel information can be determined in advance.

[0075] Step 103, determining the target angle from the various rotation angles based on the set of people that can be shot at the various rotation angles, and controlling the second camera to shoot at the target angle.

[0076] Specifically, the target angle is the optimal rotation angle of the second camera. According to the set of people that can be shot at each rotation angle of the second camera, at any time, the optimal rotation angle can be determined from the various rotation angles, and the second camera is controlled to shoot at the optimal rotation angle.

[0077] The shooting method provided by the embodiment of the application determines the optimal rotation angle of the low point camera by determining the set of people that can be shot at different rotation angles of the low point camera, thereby controlling the low point camera to shoot at the optimal rotation angle, avoiding the missed shooting of the personnel face photo caused by too many blind spots or inappropriate camera angles, and improving the overall crowd face capture rate.

[0078] In some embodiments, the target angle is determined from the various rotation angles based on the set of people that can be shot at the various rotation angles, comprising:

[0079] Based on at least one of the position, speed, face acquisition rate and face library comparison degree of each person in the set of people to be shot, a shooting weight of each person is determined;

[0080] Based on the shooting weight of each person in the set of people that can be shot at the various rotation angles, a shooting score of the various rotation angles is determined.

[0081] The target angle is determined from the various rotation angles based on the shooting score of the various rotation angles.

[0082] Specifically, the process of shooting is advancing with time, and at different times, the calculation method of selecting the optimal rotation angle of the second camera according to the set of people that can be shot at each rotation angle of the second camera can be different. The selection of the optimal rotation angle can be realized according to the shooting score of the second camera at different rotation angles, and the shooting score at different rotation angles depends on the shooting weight of each person in the set of people that can be shot.

[0083] Optionally, the calculation method of the optimal rotation angle of the future time can be corrected according to the comparison result of the frontal face snapshot of the image of the second camera at the historical time at the optimal rotation angle.

[0084] The personnel information obtained by the high point camera and the low point camera, including but not limited to position, speed, face acquisition rate, is used to determine the shooting weight of each personnel. Further, since the shooting is often continuous, the shooting weight of the personnel at the future time can also be determined according to the comparison coincidence degree of the face library at the historical time.

[0085] In the snapshot area covered by the high point camera array, the closer the personnel is to the center of the snapshot area, the more low point cameras can be snapped, the higher the possibility of being snapped to the frontal face, the lower the shooting priority, and the lower the shooting weight. Conversely, the farther the personnel is from the center, the higher the shooting priority, and the higher the shooting weight.

[0086] In the snapshot area covered by the high point camera array, the faster the personnel moves, the more likely it is to be missed, the higher the shooting priority, and the higher the shooting weight. Conversely, the slower the personnel moves, the more likely it is to be snapped, the lower the shooting priority, and the lower the shooting weight.

[0087] In the snapshot area covered by the high point camera array, the face acquisition rate is affected by the frontal face snapshot rate and the number of snapable low point cameras. In the image of the person snapped by the low point camera, the head region is not necessarily a face, the face may be facing away or sideways to the camera, the more images containing no face are snapped, the lower the frontal face snapshot rate, and the lower the face acquisition rate. At the same time, the more low point cameras that can snap the specified personnel, the higher the face acquisition rate. The higher the face acquisition rate, the lower the shooting priority, and the lower the shooting weight. Conversely, the lower the face acquisition rate, the higher the shooting priority, and the higher the shooting weight.

[0088] Since the shooting is continuous, the snapped personnel image will be compared with the face library at a fixed time interval, and there can be one or more comparison results. If there are multiple comparison results, the more comparison results, the less clear the face of the snapped face image, and the higher the subsequent shooting weight; the fewer comparison results, the clearer the frontal face of the snapped face image, and the lower the subsequent shooting weight. If there is only one comparison result, it means that the snapped face is the face with the highest matching degree, i.e., the frontal face is snapped. Optionally, for the personnel whose frontal face has been snapped, the personnel can not be snapped within a certain time period in the snapshot area covered by the current high point camera array.

[0089] Based on the shooting weight of each personnel in the set of snapable personnel at each rotation angle, the shooting points of each rotation angle can be determined, and thus the target angle can be determined from each rotation angle.

[0090] The photographing method provided by the embodiments of the present application dynamically calculates the personnel capture score of each rotation angle of the low point camera according to the position, speed, face acquisition rate and face library comparison coincidence degree of the personnel, so as to determine the optimal rotation angle of the low point camera, and effectively improves the capture rate of the front face.

[0091] In some embodiments, the photographing weight of each personnel in the set of personnel to be photographed is determined based on at least one of the position, speed, face acquisition rate and face library comparison coincidence degree of the personnel.

[0092] The distance between the personnel and the center point of the photographing area corresponding to the first camera is determined based on the position of the personnel and the position of the center point.

[0093] The photographing weight of the personnel is determined based on the distance and the number of the second cameras corresponding to the set of photographable personnel in which the personnel exists.

[0094] Specifically, in the photographing area of the high point camera, the photographing weight of the personnel is mainly affected by the position of the personnel from the center point of the photographing area of the high point camera. The farther the distance, the less likely the personnel is captured, and the higher the photographing weight.

[0095] The distance between the personnel and the center point of the photographing area of the high point camera is determined based on the position of the personnel and the position of the center point. Optionally, the distance is the actual distance between the personnel and the center point. Optionally, the actual distance between the personnel closest to the center point and the center point is taken as a unit distance, and the distances between the other personnel and the center point are determined.

[0096] The photographing weight of the personnel is determined based on the distance calculated in the foregoing step and the number of the second cameras corresponding to the set of photographable personnel in which the personnel exists. In the foregoing step, the set of photographable personnel of the low point camera at any rotation angle at any moment is determined. If the personnel is in the set of photographable personnel of a low point camera, it is indicated that the low point camera is the photographable low point camera of the personnel, so that the number of the photographable low point cameras corresponding to the personnel is determined.

[0097] Case 1: The default rotation angle of the low point camera before being started is the initial 0 degree angle (see Figure 2 ). Before starting the low point camera, the set of photographable personnel of the low point camera at each rotation angle is calculated in advance to obtain a set of rotatable angles. Each rotation angle includes a different set of photographable personnel, and the number of the photographable personnel is m.

[0098] The capture score of the personnel is set as r. When the personnel is in the initial position, the capture score of the personnel in the high point area is mainly affected by the position of the personnel from the center point of the high point capture area. The farther the distance, the more likely the personnel is to leave the current high point area.

[0099] The first weight is set as a. The acquisition manner of the first weight depends on the distance l of the person from the center point of the high point area of the camera (optionally, the distance between the person closest to the center point and the center point is taken as a unit distance, and the minimum distance is 1). In combination with the number v of low point cameras that can capture the person, the following can be obtained:

[0100]

[0101] It can be known from the above formula that the farther the person is from the center point of the high point area, the fewer the number of cameras that can capture the person, and the greater the first weight of the person. Conversely, the closer the person is to the center point, the more the number of cameras that can capture the person, and the smaller the first weight, which is even less than 1 and approaches 0 infinitely.

[0102] The weighted capture score of the person can be obtained through the first weight and the capture score r of the person. Therefore, the total person score that can be captured by the low point camera when the horizontal angle is rotated by a and the pitch angle is rotated by β can be pre-calculated as:

[0103]

[0104] wherein r i is the weighted capture score of the i-th person. Given that there are n rotation angles that can capture the person's head, the set of total person scores that can be captured under multiple rotation angles is:

[0105]

[0106] The maximum value is selected from the above set, and the rotation angle corresponding to the maximum value is the initial optimal rotation angle at T1. Optionally, if there are multiple maximum values, the rotation angle corresponding to the set of people that can be captured with the largest number of people is taken as the initial optimal rotation angle.

[0107] The photographing method provided in the embodiments of the present application determines the photographing weight of the person by the distance of the person from the center point of the high point camera photographing area and the set of low point cameras corresponding to the person, thereby further improving the face capture rate of the crowd.

[0108] In some embodiments, the photographing weight of the person is determined based on the distance and the number of second cameras corresponding to the set of people that can be captured.

[0109] The first weight of the person is determined based on the distance and the number of second cameras corresponding to the set of people that can be captured.

[0110] The second weight of the person is determined based on at least one of the speed of the person, the face acquisition rate, and the face library comparison degree of conformity.

[0111] determine a shooting weight of the person based on the first weight and the second weight of the person.

[0112] Specifically, according to the distance between the person and the center point of the high-point camera shooting area, in combination with the number of low-point cameras corresponding to the person, the first weight of the person can be determined.

[0113] At the initial shooting moment, the first weight can be taken as the shooting weight of the person, and the optimal rotation angle of the low-point camera is calculated, and shooting is performed at the optimal rotation angle. Compare the photographed image of the person with the face library. If there are multiple comparison results, it means that the accuracy of the optimal rotation angle calculated by taking only the first weight as the shooting weight is not enough.

[0114] Therefore, in the subsequent shooting process, the second weight of the person should be determined in combination with the first weight, at least one of the speed of the person, the person acquisition rate and the face library comparison compliance.

[0115] determine a shooting weight of the person based on the first weight and the second weight of the person.

[0116] Case 2: Because the personnel in the same high-point area are constantly changing, after the snapshot at T1 is completed, the personnel information in the current area needs to be acquired again through the high-point camera. Optionally, personnel who have left the area can be removed, and new personnel who have entered the area can be added to reduce the calculation complexity. For personnel who have not left the area and have acquired the snapshot result at T1, in combination with the personnel information acquired by the high-point camera before, the snapped face can be associated with the corresponding personnel information. Compare the snapped face with the face in the population library. If there is only one archive in the matching result, it proves that the face snapped this time is the most valuable face, and the personnel can be excluded from the target group in the subsequent snapshot.

[0117] If there are multiple face archives matched, the shooting weight of the personnel needs to be adjusted, specifically:

[0118] Set the second weight of the personnel as d, d is the result value after the face in the snapped face and the face in the population library is compared, that is, the face library comparison compliance, set the maximum number of TOP that can be compared as p (a pre-set threshold, that is, the maximum number of face comparisons), the number of faces compared is q, and the initial value of the face not snapped is p by default. Then, the following can be calculated:

[0119]

[0120] As shown in the formula above, when a captured face is compared with faces in the face database, the fewer matching faces there are, the clearer and more accurate the captured face image is. This results in a lower match rate with the face database, and a lower need to recapture the person's face. Similarly, when too many matching faces are found, the match rate with the face database is high, indicating a quality issue with the captured face image. Although the face was captured, it may be necessary to recapture it. Furthermore, based on the results of each captured face comparison, if the match rate with the face database is high and the number of matched files is less than before, then the face previously associated with the person's information will be replaced.

[0121] After excluding those whose faces were captured at time T1, and considering the camera's snapshot data from time T1, before time T2 begins, it is necessary to calculate the individual's score before time T2 by combining the two states. Specifically, the weighted shooting weight for each individual is determined by combining 'a' and 'd', and the individual's actual score is determined based on the weighted shooting weight.

[0122] Scenario 3: Since the movement speed of a person within a high-point area also affects the capture rate, a second weight of b is assigned to this person, where b is the person's movement speed within the high-point area. b is obtained as follows: when T... n-1 Ideally, the person should be captured at the designated time, but if the person does not appear at the predetermined location, then T is obtained. v Time and T n-1 The actual distance s between people at any given time can be calculated as follows:

[0123]

[0124] As can be seen from the above, the faster a person moves, the easier it is for them to be missed in a snapshot. Optionally, based on the person's direction of movement, speed of movement, and the time it takes for the camera to rotate, the position the person may reach when the camera finishes rotating at a low point can be calculated in advance, and the position for the next snapshot will be based on the pre-calculated position.

[0125] Scenario 4: The face acquisition rate of personnel also affects the capture rate. Therefore, a second weight, c, is assigned to this personnel, where c is the face acquisition rate, representing the number of consecutive face images that are not captured. c is obtained as the product of the number of times the personnel are photographed (t) and the cumulative number of cameras capable of capturing low-point images, as detailed below:

[0126] c = t(v1 + v2 + ... + v) t )

[0127] Among them, v n This indicates the number of low-point cameras that can capture images of a specified person during the nth shooting session.

[0128] If a person's walking route is with their back to the camera, or if their face is deliberately obscured, and the face cannot be correctly captured multiple times in a row, and multiple cameras cannot capture the face even when shooting together, then the priority for capturing the person's face image is higher.

[0129] The number of shots t and the number of cameras v are continuously accumulated. For example, if the first capture attempt has 1 shot and 1 camera captures the image, the final value is 1. If the second capture attempt has 2 shots and 3 cameras can capture the image, the final value is 2 × (1 + 3) = 8.

[0130] In practice, the first weight 'a' is given priority, and the second weight can be one or more of b, c, and d. The order of consideration for multiple weights of b, c, and d is not fixed.

[0131] Optionally, considering a, b, c, and d simultaneously, and setting the initial score of the person to r, then the actual capture score of the person before time T2 is:

[0132]

[0133] The shooting method provided in this application prioritizes the position of each person in the set of people to be photographed to calculate the first weight. If the accuracy of the optimal rotation angle of the low-point camera calculated using the first weight as the shooting weight is insufficient, the method further combines at least one of the person's speed, person acquisition rate, and face database comparison conformity to determine the second weight of the person. The shooting weight of the person is determined by combining the first weight and the second weight, which improves the face frontal capture rate while reducing the computational complexity.

[0134] In some embodiments, the step of determining the face acquisition rate includes:

[0135] Obtain the number of second cameras that historically photographed the person as part of the set of people who could be photographed, and the number of times the person was photographed historically as part of the set of people who could be photographed;

[0136] The face acquisition rate of the person is determined based on the number of the second cameras and the number of shots taken.

[0137] Specifically, since the filming process is often continuous in time, the capture rate varies at different times.

[0138] The face acquisition rate can be calculated as follows: First, obtain the number of available second cameras corresponding to each person in history, and the number of times each person will be photographed. Then, combine these two data points to determine the face acquisition rate. There is a positive correlation between the face acquisition rate, the number of available second cameras corresponding to each person, and the number of times each person will be photographed.

[0139] The shooting method provided in this application takes into account the face acquisition rate during the calculation of the shooting weight of people, and focuses more on people with low face acquisition rates, thereby further improving the capture rate of faces in the crowd.

[0140] In some embodiments, the step of determining each rotation angle includes:

[0141] According to the shooting weight from high to low, the corresponding rotation angle is determined with the position of each person in the group of people to be photographed as the shooting center.

[0142] Specifically, after calculating the shooting weight of each person, each person in the set of people to be photographed is taken as the center point of the low-point camera according to the shooting weight of each person in the set of people to be photographed from high to low. The set of people to be photographed is obtained with this center point, and the total shooting weight of people in the capture area when this person is taken as the center point is calculated.

[0143] The person with the highest total shooting weight is selected as the capture center point for the next moment, thus determining the initial rotation angle of the low-point camera at the next moment. Optionally, if there are multiple maximum total shooting weights, the rotation angle with the highest number of people that can be captured is used as the initial rotation angle for the next moment.

[0144] Specifically, after calculating the actual capture score of each person in the area before time T2, the people are sorted from highest to lowest capture score, in the order from 1 to w. Each person in the sorted order is taken as the center point for the low-point camera to capture at the next time. The number of people that the low-point camera can capture is obtained from this center point. The total score of people in the capture area when a single person is taken as the center point can be calculated:

[0145]

[0146] The set of total points for individuals sorted from 1 to w is:

[0147]

[0148] The maximum value is selected from the above set. If there are multiple maximum values, the one with the most people captured is given priority. The corresponding rotation angle is then determined as the initial rotation angle at time T2.

[0149] The shooting method provided in this application selects the rotation angle of the low-point camera based on the shooting weight of the people, thereby selecting the optimal rotation angle of the low-point camera and improving the capture rate of the faces of the crowd.

[0150] In some embodiments, it also includes:

[0151] Acquire a second image captured by the second camera at the target angle;

[0152] Based on the second image, the people who have already been photographed are determined from the set of people who can be photographed from the target angle, and the first camera is controlled again to take a new first image;

[0153] The process of determining the set of people to be photographed based on the first image captured by the first camera includes:

[0154] Based on the new first image, a set of people to be photographed, after filtering out the people already photographed, is determined.

[0155] Specifically, the low-point camera is controlled to take a picture at an optimal rotation angle, acquiring a second image captured by the low-point camera at the optimal rotation angle. Based on the second image, the people already photographed are determined from the set of people that can be photographed at the optimal rotation angle. Optionally, the people already photographed are determined from the set of people that can be photographed at the optimal rotation angle based on the comparison result between the second image and a face database. Optionally, the photographed person is the person whose face database comparison result shows only one person. When the face database comparison result shows only one person, it means that the person's face has been captured, and in subsequent shooting processes, the person can be filtered out from the picture.

[0156] The system controls a first camera to capture a new first image, and filters out people already photographed from a set of people identified in the new first image. The set of people to be photographed based on the first image captured by the first camera can be the set of people to be photographed after filtering out those already photographed.

[0157] Specifically, the presence of people within the entire high-point area is a continuous process, one that can be filmed from T1 to Tn. The initial set of people at time T1 is known to be U. T1 As time changes, the number of people gathered in the high-point area is constantly changing, denoted as U. Tn Then, as time changes, the complete set of data becomes:

[0158] (U T1 U T2 ,…,U Tn )

[0159] It can be seen that, as time changes, the total header data that should be obtained at time Tn is:

[0160] U all =U T1 ∪U T2 ∪…∪U Tn

[0161] At the initial time T1, the default set of captured faces is P. T1Then, at the current time T1, the number of faces that still need to be captured is P. T1 In the supplementary collection of all facial data that need to be captured At the next moment T2, the set of captured faces is P. T2 The remaining faces that need to be captured are People captured at time T1 may also exist. Within this range, the actual remaining captured faces at time T2 are:

[0162]

[0163] Therefore, the remaining faces captured at time Tn are:

[0164]

[0165] The above steps calculate the number of faces to be captured, assuming that rotating to the corresponding position guarantees a 100% capture rate. However, in reality, people are in various states, leading to inaccurate capture data. Therefore, it is necessary to differentiate the capture weight for people in different states.

[0166] The shooting method provided in this application filter out people already photographed during the shooting process, thereby improving the face capture rate of crowds while reducing the computational complexity.

[0167] Optionally, the calculation method for the optimal rotation angle in future moments can be corrected based on the comparison results of frontal face captures of images taken by the second camera at the optimal rotation angle at historical moments.

[0168] Specifically, after rotating and capturing images at time T2, it is necessary to exclude individuals whose faces cannot be captured. These individuals are those whose faces cannot be captured after multiple attempts. It is assumed that with the current camera installation position and camera rotation angle, it is impossible to capture images of these individuals.

[0169] Then set the maximum number of captures threshold t. max When the number of times a person is captured is greater than or equal to the threshold, it is considered that the person's facial information cannot be obtained within the capture area covered by the current high-point camera array, no matter how the low-point camera is rotated. The person's information is recorded and placed into the next high-point area to continue capturing the person.

[0170] After the elimination is complete, repeat the entire calculation process of the camera rotation angle at time T2 to calculate the angle that the camera needs to rotate at time T3.

[0171] Furthermore, for faces not captured within the current high-point camera array's coverage area, the individual's body information is recorded. When the individual enters the next high-point visible area, a third weight should be set to e, where e is the number of high-point camera areas that have not yet captured the individual's optimal face. Therefore, this capture weight needs to be increased when calculating the individual's capture score until the individual's optimal face is captured.

[0172] The shooting method provided in this application embodiment, within the shooting area of ​​a high-point camera, leverages the linkage between the high-point camera and a low-point camera. By capturing personnel information through the high-point camera and combining it with the latitude and longitude data of the low-point camera and the camera's capture range, it can calculate whether the person's current location can be captured by the low-point camera, and calculate the required rotation angle of the camera to capture the face. For different personnel states within the area, by combining information such as the person's location, speed, face acquisition rate, and compatibility with the population database, it dynamically calculates the personnel capture integral that can be obtained for each rotation angle of the low-point camera, thereby determining the optimal rotation position at the current moment and improving the face capture rate of the crowd.

[0173] The shooting device provided by the present invention is described below. The shooting device described below can be referred to in correspondence with the shooting method described above.

[0174] Figure 3 This is a schematic diagram of the imaging device provided in the embodiments of this application, as shown below. Figure 3 As shown, the device includes at least:

[0175] The first determining unit 301 is used to determine the set of people to be photographed and the personnel information of each person in the set of people to be photographed based on the first image captured by the first camera.

[0176] The second determining unit 302 is used to determine, based on the camera parameters of the second camera and the position information in the personnel information, the set of persons that can be photographed by the second camera at various rotation angles from the set of persons to be photographed, wherein the setting positions of the first camera and the second camera are different;

[0177] The third determining unit 303 is used to determine the target angle from the set of people that can be photographed at each rotation angle, and control the second camera to take pictures at the target angle.

[0178] In some embodiments, the third determining unit 303 is specifically used for:

[0179] The shooting weight of each person is determined based on at least one of the following: position, speed, face acquisition rate, and face database matching degree of each person in the set of people to be photographed.

[0180] Based on the shooting weight of each person in the set of people who can be photographed at each rotation angle, the shooting integral for each rotation angle is determined.

[0181] The target angle is determined from the shooting integrals of each rotation angle.

[0182] In some embodiments, the third determining unit 303 is specifically used for:

[0183] Based on the location of the person and the location of the center point of the shooting area corresponding to the first camera, the distance between the person and the center point is determined;

[0184] Based on the distance and the number of second cameras corresponding to the person in the set of people who can be photographed, the shooting weight of the person is determined.

[0185] In some embodiments, the third determining unit 303 is specifically used for:

[0186] Based on the distance and the number of second cameras corresponding to the person in the set of people who can be photographed, a first weight of the person is determined;

[0187] The second weight of the person is determined based on at least one of the person's speed, face acquisition rate, and face database matching degree;

[0188] The shooting weight of the person is determined based on the first weight and the second weight of the person.

[0189] In some embodiments, the step of determining the face acquisition rate includes:

[0190] Obtain the number of second cameras that historically photographed the person as part of the set of people who could be photographed, and the number of times the person was photographed historically as part of the set of people who could be photographed;

[0191] The face acquisition rate of the person is determined based on the number of the second cameras and the number of shots taken.

[0192] In some embodiments, the step of determining each rotation angle includes:

[0193] According to the shooting weight from high to low, the corresponding rotation angle is determined with the position of each person in the group of people to be photographed as the shooting center.

[0194] In some embodiments, the device further includes:

[0195] The acquisition unit is used to acquire a second image captured by the second camera at the target angle;

[0196] The fourth determining unit is used to determine the people already photographed from the set of people that can be photographed from the target angle based on the second image, and to re-control the first camera to take a new first image;

[0197] The first determining unit 301 is specifically used for:

[0198] Based on the new first image, a set of people to be photographed, after filtering out the people already photographed, is determined.

[0199] Figure 4 This is a schematic diagram of the imaging system provided in the embodiments of this application, as shown below. Figure 4 As shown, the system includes at least:

[0200] The first camera 401, the second camera 402, and any of the aforementioned shooting devices 403 are connected together.

[0201] Figure 5 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application, such as... Figure 5 As shown, the electronic device may include: a processor 501, a communication interface 502, a memory 503, and a communication bus 504, wherein the processor 501, the communication interface 502, and the memory 503 communicate with each other via the communication bus 504. The processor 501 can call logical instructions in the memory 503 to execute a shooting method, including:

[0202] Based on the first image captured by the first camera, determine the set of people to be photographed and the personnel information of each person in the set of people to be photographed;

[0203] Based on the camera parameters of the second camera and the location information in the personnel information, the set of personnel that can be photographed by the second camera at various rotation angles is determined from the set of personnel to be photographed. The first camera and the second camera are set in different positions.

[0204] Based on the set of people who can be photographed at each rotation angle, the target angle is determined from the rotation angles, and the second camera is controlled to take pictures at the target angle.

[0205] Furthermore, the logical instructions in the aforementioned memory 503 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0206] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program, the computer program being able to be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer being able to execute the imaging methods provided by the above methods, including:

[0207] Based on the first image captured by the first camera, determine the set of people to be photographed and the personnel information of each person in the set of people to be photographed;

[0208] Based on the camera parameters of the second camera and the location information in the personnel information, the set of personnel that can be photographed by the second camera at various rotation angles is determined from the set of personnel to be photographed. The first camera and the second camera are set in different positions.

[0209] Based on the set of people who can be photographed at each rotation angle, the target angle is determined from the rotation angles, and the second camera is controlled to take pictures at the target angle.

[0210] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the imaging methods provided by the above methods, including:

[0211] Based on the first image captured by the first camera, determine the set of people to be photographed and the personnel information of each person in the set of people to be photographed;

[0212] Based on the camera parameters of the second camera and the location information in the personnel information, the set of personnel that can be photographed by the second camera at various rotation angles is determined from the set of personnel to be photographed. The first camera and the second camera are set in different positions.

[0213] Based on the set of people who can be photographed at each rotation angle, the target angle is determined from the rotation angles, and the second camera is controlled to take pictures at the target angle.

[0214] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0215] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0216] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A shooting method, characterized in that, include: Based on the first image captured by the first camera, determine the set of people to be photographed and the personnel information of each person in the set of people to be photographed; Based on the camera parameters of the second camera and the location information in the personnel information, the set of personnel that can be photographed by the second camera at various rotation angles is determined from the set of personnel to be photographed. The first camera and the second camera are set in different positions. Based on the set of people who can be photographed at each rotation angle, the target angle is determined from the rotation angles, and the second camera is controlled to take pictures at the target angle.

2. The shooting method according to claim 1, characterized in that, The set of people to be photographed based on each rotation angle, determining the target angle from each rotation angle, includes: The shooting weight of each person is determined based on at least one of the following: position, speed, face acquisition rate, and face database matching degree of each person in the set of people to be photographed. Based on the shooting weight of each person in the set of people who can be photographed at each rotation angle, the shooting integral for each rotation angle is determined. The target angle is determined from the shooting integrals of each rotation angle.

3. The shooting method according to claim 2, characterized in that, The step of determining the shooting weight of each person based on at least one of the following: position, speed, face acquisition rate, and face database matching degree of each person in the set of people to be photographed, includes: Based on the location of the person and the location of the center point of the shooting area corresponding to the first camera, the distance between the person and the center point is determined; Based on the distance and the number of second cameras corresponding to the person in the set of people who can be photographed, the shooting weight of the person is determined.

4. The shooting method according to claim 3, characterized in that, The determination of the shooting weight of a person based on the distance and the number of second cameras corresponding to the person in the set of people who can be photographed includes: Based on the distance and the number of second cameras corresponding to the person in the set of people who can be photographed, a first weight of the person is determined; The second weight of the person is determined based on at least one of the person's speed, face acquisition rate, and face database matching degree; The shooting weight of the person is determined based on the first weight and the second weight of the person.

5. The shooting method according to claim 2, characterized in that, The steps for determining the face acquisition rate include: Obtain the number of second cameras that historically photographed the person as part of the set of people who could be photographed, and the number of times the person was photographed historically as part of the set of people who could be photographed; The face acquisition rate of the person is determined based on the number of the second cameras and the number of shots taken.

6. The shooting method according to claim 2, characterized in that, The steps for determining each rotation angle include: According to the shooting weight from high to low, the corresponding rotation angle is determined with the position of each person in the group of people to be photographed as the shooting center.

7. The shooting method according to any one of claims 1 to 6, characterized in that, Also includes: Acquire a second image captured by the second camera at the target angle; Based on the second image, the people who have already been photographed are determined from the set of people who can be photographed from the target angle, and the first camera is controlled again to take a new first image; The process of determining the set of people to be photographed based on the first image captured by the first camera includes: Based on the new first image, a set of people to be photographed, after filtering out the people already photographed, is determined.

8. A shooting device, characterized in that, include: The first determining unit is used to determine the set of people to be photographed and the personnel information of each person in the set of people to be photographed based on the first image captured by the first camera. The second determining unit is used to determine, based on the camera parameters of the second camera and the position information in the personnel information, the set of persons that can be photographed by the second camera at various rotation angles from the set of persons to be photographed, wherein the setting positions of the first camera and the second camera are different; The third determining unit is used to determine the target angle from the set of people that can be photographed at each rotation angle, and control the second camera to take pictures at the target angle.

9. A shooting system, characterized in that, It includes a first camera and a second camera connected together, as well as the device as described in claim 8.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the shooting method as described in any one of claims 1 to 7.

Images