Position point determination method and apparatus, electronic device, and storage medium

Abstract

The present disclosure relates to a position point determination method and device, electronic equipment and storage medium, the method comprising: obtaining a display region corresponding to a display image and a touch region, the touch region being obtained in response to a touch operation on the display image; based on the display region and the touch region, obtaining a distance weight corresponding to each feature extraction sub-region included in the display region, the distance weight representing a distance relationship between the each feature extraction sub-region and the target position point; performing feature extraction on the each feature extraction sub-region to obtain a feature vector corresponding to the each feature extraction sub-region; and determining a target position point based on the distance weight corresponding to the each feature extraction sub-region and the feature vector corresponding to the each feature extraction sub-region. The method of the present disclosure can make the determined position point more in line with the actual needs of the user, thereby improving the accuracy of position point determination.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of computer, and particularly relates to a position point determination method and device, electronic equipment and storage medium. BACKGROUND

[0002] In the use of electronic equipment, a user can interact with a screen through a touch manner, so as to select a position point in the screen. However, when the user selects the position point through the touch manner, there is a problem of inaccurate selection of the position point. SUMMARY

[0003] To overcome the problems in the prior art, the present disclosure provides a position point determination method and device, electronic equipment and storage medium.

[0004] According to a first aspect of an embodiment of the present disclosure, a position point determination method is provided, and the method comprises:

[0005] obtaining a display region corresponding to a display image and a touch region, the touch region being obtained in response to a touch operation on the display image;

[0006] based on the touch region, obtaining distance weights corresponding to each feature extraction sub-region included in the display region, the distance weights representing distance relationships between the each feature extraction sub-region and a center point of the touch region;

[0007] performing feature extraction on the each feature extraction sub-region to obtain a feature vector corresponding to the each feature extraction sub-region;

[0008] based on the distance weights corresponding to the each feature extraction sub-region and the feature vector corresponding to the each feature extraction sub-region, determining a target position point.

[0009] In some embodiments, the obtaining, based on the touch region, of the distance weights corresponding to the each feature extraction sub-region included in the display region comprises:

[0010] obtaining a first distance between the center point of the touch region and a center point of a candidate feature extraction sub-region, and obtaining a second distance between the center point of the touch region and a farthest boundary of the touch region, the candidate feature extraction sub-region being any one of the each feature extraction sub-region;

[0011] based on the first distance and the second distance, determining a distance weight corresponding to the candidate feature extraction sub-region.

[0012] In some embodiments, the determining, based on the first distance and the second distance, of the distance weight corresponding to the candidate feature extraction sub-region comprises:

[0013] obtaining a ratio of the first distance and the second distance;

[0014] determining, according to the ratio, a distance weight corresponding to the candidate feature extraction sub-region, the distance weight being inversely related to the ratio;

[0015] The method further includes determining a target position point based on the distance weight corresponding to each feature extraction sub-region and the feature vector corresponding to each feature extraction sub-region.

[0016] The method further includes pre-processing the feature vector corresponding to each feature extraction sub-region to obtain a pre-processed feature value corresponding to each feature extraction sub-region.

[0017] The method further includes obtaining a product of the pre-processed feature value corresponding to each feature extraction sub-region and the distance weight corresponding to each feature extraction sub-region.

[0018] The method further includes determining a center point of a feature extraction sub-region with the largest product as the target position point.

[0019] In some embodiments, the method includes:

[0020] The method further includes determining the feature extraction sub-regions around the center point of the touch region as feature extraction windows with a target size.

[0021] In some embodiments, the feature extraction on the feature extraction sub-regions includes:

[0022] The method further includes obtaining a target scene corresponding to the display image.

[0023] The method further includes obtaining a target feature extraction algorithm corresponding to the target scene.

[0024] The method further includes performing feature extraction on the feature extraction sub-regions according to the target feature extraction algorithm.

[0025] In some embodiments, the target scene includes a plurality of preset scenes, and the feature extraction on the feature extraction sub-regions according to the target feature extraction algorithm includes:

[0026] The method further includes performing feature extraction on a candidate feature extraction sub-region according to a target feature extraction algorithm corresponding to each preset scene to obtain a sub-feature vector of the candidate feature extraction sub-region under each target feature extraction algorithm, the candidate feature extraction sub-region being any one of the plurality of feature extraction sub-regions.

[0027] The method further includes determining a target position point based on the distance weight corresponding to each feature extraction sub-region and the feature vector corresponding to each feature extraction sub-region.

[0028] The sub-feature vectors corresponding to the candidate feature extraction sub-region under each target feature extraction algorithm are preprocessed respectively to obtain preprocessed sub-feature values corresponding to the candidate feature extraction sub-region under each target feature extraction algorithm.

[0029] The sub-feature values corresponding to the candidate feature extraction sub-region under each target feature extraction algorithm are obtained, and the sub-feature value under one target feature extraction algorithm is the product of the distance weight corresponding to the candidate feature extraction sub-region and the preprocessed sub-feature value under the corresponding target feature extraction algorithm.

[0030] The sub-feature values corresponding to the candidate feature extraction sub-region under each target feature extraction algorithm are weighted and summed according to the weight values corresponding to the target feature extraction algorithms to obtain the comprehensive feature values corresponding to the candidate feature extraction sub-region, and the weight value corresponding to each feature extraction algorithm represents the confidence of each target feature extraction algorithm under the corresponding preset scene.

[0031] The target position point is determined based on the comprehensive feature values corresponding to each candidate feature extraction sub-region.

[0032] According to a second aspect of the embodiments of the present disclosure, a position point determination apparatus is provided, which comprises:

[0033] A region determination module configured to obtain a display region corresponding to a display image and a touch region, the touch region being obtained in response to a touch operation on the display image;

[0034] A distance weight determination module configured to obtain distance weights corresponding to each feature extraction sub-region included in the display region based on the touch region, the distance weights representing distance relationships between the feature extraction sub-regions and a center point of the touch region.

[0035] A feature extraction module configured to perform feature extraction on the feature extraction sub-regions to obtain feature vectors corresponding to the feature extraction sub-regions.

[0036] A target position point determination module configured to determine a target position point based on the distance weights corresponding to the feature extraction sub-regions and the feature vectors corresponding to the feature extraction sub-regions.

[0037] According to a third aspect of the embodiments of the present disclosure, a computer readable storage medium is provided, which stores computer program instructions, and the program instructions are executed by a processor to implement the steps of the position point determination method provided in the first aspect of the present disclosure.

[0038] According to a fourth aspect of the embodiments of the present disclosure, an electronic device is provided, comprising: a memory having a computer program stored thereon; and a processor configured to execute the computer program in the memory to implement the steps of the position point determination method according to the first aspect of the present disclosure.

[0039] The technical solutions provided by the embodiments of the present disclosure can have the following beneficial effects: since the distance weight and the feature vector corresponding to each feature extraction sub-region are considered simultaneously in the process of determining the target position point, and since the distance weight represents the distance relationship between each feature extraction sub-region and the center point of the touch region, on the one hand, the possibility that the center point of the user's touch region is the target position point is considered, and on the other hand, the possibility that the user is more likely to select a meaningful point is also considered, so that the target position point determined by combining the two aspects can be more in line with the actual needs of the user, thereby improving the accuracy of the position point determination.

[0040] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure BRIEF DESCRIPTION OF DRAWINGS

[0041] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the present disclosure and, together with the specification, serve to explain the principles of the present disclosure.

[0042] Figure 1 is a flowchart of a position point determination method according to an exemplary embodiment.

[0043] Figure 2 is a flowchart of another position point determination method according to an exemplary embodiment.

[0044] Figure 3 is a flowchart of another position point determination method according to an exemplary embodiment.

[0045] Figure 4 is a flowchart of another position point determination method according to an exemplary embodiment.

[0046] Figure 5 is a flowchart of another position point determination method according to an exemplary embodiment.

[0047] Figure 6 is a schematic diagram of a comprehensive feature value determination process according to an exemplary embodiment.

[0048] Figure 7 is a structural block diagram of a position point determination device according to an exemplary embodiment.

[0049] Figure 8is a block diagram of an electronic device according to an example embodiment. DETAILED DESCRIPTION

[0050] The example embodiments will be described in detail herein with reference to the accompanying drawings. In the following description, unless otherwise indicated, like numbers in the different drawings represent the same or similar elements. The following detailed description does not, therefore, limit the scope of the disclosure, but the scope of the disclosure is defined by the appended claims.

[0051] During the use of the electronic device, there are some visual-based user interaction scenarios in which the user can interact with the screen by touch (e.g., finger touch or stylus touch, etc.) to select a position point in the screen. For example, when the user performs 3D distance measurement using the electronic device, the user can select two position points for which the distance is to be measured by touch. For another example, when the camera is focused by touch, the user can select the focus point by touch. For yet another example, in a game application, the user can control the game character to move to a target point by clicking a point on the screen by touch.

[0052] However, when the user selects a position point by touch, the touch object (finger or stylus, etc.) has a certain contact area with the screen, so the touch actually selects a region, not an accurate point. In the related art, the center point of the touch contact region is directly selected as the selected position point, which has the problem of inaccurate position point selection. Therefore, how to improve the accuracy of position point determination is a problem to be solved.

[0053] Figure 1 is a flowchart of a position point determination method according to an example embodiment. The method can be applied to an electronic device, such as a mobile phone, a notebook, a tablet computer, a desktop computer, a smart television, a wearable device, a smart robot, a vehicle terminal, etc. The method comprises:

[0054] S110, obtaining a display region corresponding to a display image and a touch region, the touch region being obtained in response to a touch operation on the display image.

[0055] The display image can be understood as an image displayed on the screen of the electronic device. In the embodiments of the present disclosure, the display image is not specifically limited, and the corresponding display image can be displayed according to the actual scene, such as a map image, a face image, and a game image, etc. In some scenarios, it can also be a solid color image, such as a black image.

[0056] In combination with the foregoing, there is a need for a user to select a position point in a screen by touch. When the user touches the screen, i.e., when the user initiates a touch operation on the displayed image, a touch region corresponding to the touch position of the user can be obtained. It can be understood that the touch region can be a part of the display region.

[0057] In the embodiments of the present disclosure, the touch region can be determined in various ways.

[0058] In some embodiments, different touch regions have different touch forces. A touch region that meets a preset touch force threshold can be determined as the touch region.

[0059] In other embodiments, different touch region sizes can be set according to the touch object. For example, a finger touch and a stylus touch can correspond to different touch region sizes. Therefore, the touch regions corresponding to different touch objects can be set according to experience. When a finger touch is performed, a region within a first range with the center of the touch surface of the finger touch as the center can be determined as the touch region. When a stylus touch is performed, a region within a first range with the center of the touch surface of the stylus touch as the center can be determined as the touch region.

[0060] In yet other embodiments, the touch surface can be directly determined as the touch region.

[0061] In addition, it can be understood that the display image has a predetermined display range in the screen, which is the display region corresponding to the display image.

[0062] In some embodiments, the display region corresponding to the display image and the touch region can be represented in units of pixels. In this case, the display region corresponding to the display image and the touch region can be converted according to the screen pixels of the electronic device.

[0063] In S120, distance weights corresponding to each feature extraction sub-region included in the display region are obtained based on the touch region. The distance weights represent the distance relationship between each feature extraction sub-region and the center point of the touch region.

[0064] In the embodiments of the present disclosure, the display region can be divided to obtain different feature extraction sub-regions. Then, image feature extraction is performed on each feature extraction sub-region to obtain a feature vector corresponding to each feature extraction sub-region.

[0065] In the embodiments of the present disclosure, the display region can be divided to obtain different feature extraction sub-regions. Then, image feature extraction is performed on each feature extraction sub-region to obtain a feature vector corresponding to each feature extraction sub-region.

[0066] In some embodiments, the feature extraction sub-regions can be determined from the entire display region, and it can be understood that even if the feature extraction sub-regions are determined from the entire display region, due to the limitation of the distance weight, the finally determined target position point is still more likely to fall in the touch region, thereby ensuring the accuracy of the determination of the target position point.

[0067] In addition, in some cases, when the size of the feature extraction sub-region selected is greater than 1 minimum region unit, for example, assuming that the minimum region unit is a pixel, when the size of the feature extraction sub-region selected is 3*3 pixels, the feature extraction sub-region can exceed the touch region, and therefore, the feature extraction sub-region can exist partially in the display region, and it is considered that the target point expected by the user is more likely to be in the touch region. Therefore, in other embodiments, at least part of the feature extraction sub-region exists in the touch region, that is, the feature extraction sub-region and the touch region have an intersection, so that it is not necessary to determine the target position point from the entire display region, thereby reducing the number of feature extraction sub-regions, and the accuracy of the determination of the target position point can be further improved.

[0068] In the embodiments of the present disclosure, after obtaining the touch region and the display region, the distance weight corresponding to each feature extraction sub-region can be obtained based on various coordinate information of the touch region. The distance weight represents the distance relationship between each feature extraction sub-region and the center point of the touch region.

[0069] S130, performing feature extraction on each feature extraction sub-region to obtain a feature vector corresponding to each feature extraction sub-region.

[0070] In the embodiments of the present disclosure, the feature vector corresponding to the feature extraction sub-region can reflect the meaning of the image existing in the feature extraction sub-region. For example, whether it is an image content more expected by the user, whether it is an image with special or jump content, and what meaning it specifically reflects can be determined according to the need by selecting a corresponding feature extraction algorithm.

[0071] There are various extraction algorithms for performing feature extraction on the feature extraction sub-region.

[0072] In some embodiments, a convolution algorithm can be used to perform feature extraction on the feature extraction sub-region.

[0073] In other embodiments, a deep feature extraction model based on a neural network can be used to perform feature extraction on the feature extraction sub-region.

[0074] S140, determining a target position point based on the distance weight corresponding to each feature extraction sub-region and the feature vector corresponding to each feature extraction sub-region.

[0075] In the embodiments of the present disclosure, after the distance weight corresponding to each feature extraction sub-region and the feature vector corresponding to each feature extraction sub-region are obtained, the target position point can be determined based on the distance weight corresponding to each feature extraction sub-region and the feature vector corresponding to each feature extraction sub-region.

[0076] By using the above method, since the distance weight and the feature vector corresponding to each feature extraction sub-region are considered in the process of determining the target position point, and since the distance weight represents the distance relationship between each feature extraction sub-region and the center point of the touch region, on the one hand, the possible case that the center point of the touch region of the user is the target position point is considered, and on the other hand, the possible case that the user is more expected to select a meaningful point is also considered, so that the target position point determined by combining the two aspects can be more in line with the actual needs of the user, thereby improving the accuracy of the position point determination.

[0077] Figure 2 FIG. 2 is a flowchart of another position point determination method according to an example embodiment, which can be applied to an electronic device, such as a mobile phone, a notebook, a tablet computer, a desktop computer, a smart television, a wearable device, a smart robot, a vehicle terminal, and the like. The method includes:

[0078] In S210, a display region corresponding to a display image and a touch region are obtained, and the touch region is obtained in response to a touch operation on the display image.

[0079] In S220, a first distance between a center point of the touch region and a center point of a candidate feature extraction sub-region, and a second distance between the center point of the touch region and a farthest boundary of the touch region are obtained, and the candidate feature extraction sub-region is any one of the feature extraction sub-regions.

[0080] In the embodiments of the present disclosure, step S220 can be performed for each feature extraction sub-region, so as to obtain the distance weight corresponding to each feature extraction sub-region.

[0081] In the embodiments of the present disclosure, for any one of the feature extraction sub-regions (i.e., the candidate feature extraction sub-region), the first distance r between the center point of the touch region and the center point of the candidate feature extraction sub-region can be obtained, and the second distance R between the center point of the touch region and the farthest boundary of the touch region can also be obtained.

[0082] The center point of the touch region can be determined according to the geometric shape of the touch region. For example, the geometric center of the touch region can be selected as the center point of the touch region.

[0083] After the center point of the touch area is determined, candidate distances between each boundary of the touch area and the center point of the touch area can be determined, and the farthest distance is selected from the candidate distances to obtain a second distance.

[0084] In S230, a distance weight corresponding to the candidate feature extraction sub-area is determined based on the first distance and the second distance.

[0085] There are multiple ways to determine the distance weight corresponding to the candidate feature extraction sub-area based on the first distance and the second distance.

[0086] In some embodiments, the ratio of the first distance and the second distance can be directly determined as the distance weight corresponding to the candidate feature extraction sub-area.

[0087] In some other embodiments, referring to Figure 3 In S230, the distance weight corresponding to the candidate feature extraction sub-area can be determined based on the first distance and the second distance, which can include the following steps:

[0088] In S231, the ratio of the first distance and the second distance is obtained.

[0089] In S232, the distance weight corresponding to the candidate feature extraction sub-area is determined according to the ratio, and the distance weight is inversely related to the ratio.

[0090] In some embodiments, the distance weight is inversely related to the ratio, and the distance weight can be expressed as 1-(r / R). That is, the greater the distance weight, the closer the feature extraction sub-area to the center point of the touch area.

[0091] It can be understood that when the distance weight is obtained based on different ways, the way to determine the target position point based on the distance weight corresponding to each feature extraction sub-area and the feature vector corresponding to each feature extraction sub-area can be different.

[0092] The method of the embodiments of the present disclosure proposes multiple ways to determine the distance weight, and improves the diversity of the position point determination method.

[0093] In S240, feature extraction is performed on each feature extraction sub-area to obtain a feature vector corresponding to each feature extraction sub-area.

[0094] In S250, the target position point is determined based on the distance weight corresponding to each feature extraction sub-area and the feature vector corresponding to each feature extraction sub-area.

[0095] As can be known from the foregoing, in some embodiments, the distance weight corresponding to the candidate feature extraction sub-area can be determined through S231 and S232, and correspondingly, please refer to Figure 3, step S250 can include the following steps:

[0096] S251, pre-process the feature vector corresponding to each feature extraction sub-region to obtain a pre-processed feature value corresponding to each feature extraction sub-region.

[0097] The feature vector is not a numerical data and cannot be directly compared, so after obtaining the feature vector corresponding to each feature extraction sub-region, pre-processing can be performed to obtain a pre-processed feature value corresponding to each feature extraction sub-region.

[0098] In some embodiments, the pre-processing of the feature vector corresponding to each feature extraction sub-region can be processed using L2 norm, wherein L2 norm processing refers to processing the sum of the squares of the absolute values of the vector elements. Thus, after L2 norm processing, numerical data can be obtained.

[0099] S252, obtain the product of the pre-processed feature value corresponding to each feature extraction sub-region and the corresponding distance weight.

[0100] S253, determine the center point of the feature extraction sub-region with the largest product as the target position point.

[0101] It can be understood that the larger the product of the pre-processed feature value corresponding to each feature extraction sub-region and the corresponding distance weight, the more the position point can meet the actual needs of the user after considering the two factors that the center point of the user touch region is the target position point and the user expects to select a meaningful point. Therefore, the center point of the feature extraction sub-region with the largest product can be determined as the target position point.

[0102] In addition, in other embodiments, in addition to determining the center point of the feature extraction sub-region with the largest product as the target position point, a pixel region of a predetermined size around the center point of the feature extraction sub-region with the largest product can also be set as the target position point according to the position point accuracy setting. That is, in some cases, the target position point can not be an absolute point, but can be composed of a pixel region of a predetermined size.

[0103] In addition, in some embodiments, before the aforementioned step S220, the step of determining each feature extraction sub-region is included, that is, the method of the embodiments of the present disclosure includes the step of: in the touch region, determining each feature extraction sub-region around the center point of the touch region with a target size as the feature extraction window.

[0104] It can be understood that in the touch region, the finally determined each feature extraction sub-region and the touch region have an overlapping part.

[0105] The target size can be set as needed, for example, 3*3, 5*5, etc.

[0106] In some embodiments, the feature extraction windows can be adjacent. In other embodiments, the feature extraction windows can also partially overlap.

[0107] It can be understood that the touch area can not be exactly filled with an integer number of feature extraction windows, resulting in a final remaining pixel area. Therefore, in the embodiments of the present disclosure, the feature extraction sub-areas can be determined from the center point of the touch area to the surrounding. Since the feature extraction sub-areas are selected in an inside-out manner, it can avoid missing a certain pixel area in the touch area.

[0108] In addition, considering that the target position point is more likely to be in the touch area, determining the feature extraction sub-areas in the touch area can reduce the number of feature extraction sub-areas and improve the efficiency of determining the target position point.

[0109] Figure 4 is a flowchart of another position point determination method according to an exemplary embodiment. The method can be applied to electronic devices, such as mobile phones, notebooks, tablets, desktop computers, smart televisions, wearable devices, smart robots, and vehicle terminals. The method includes:

[0110] S310, obtaining a display area corresponding to a display image and a touch area, the touch area being obtained in response to a touch operation on the display image.

[0111] S320, obtaining distance weights corresponding to each feature extraction sub-area included in the display area based on the touch area, the distance weights representing distance relationships between each feature extraction sub-area and a center point of the touch area.

[0112] S330, obtaining a target scene corresponding to the display image.

[0113] There are many cases for obtaining the target scene corresponding to the display image.

[0114] In some cases, the scene corresponding to the display image can be explicitly known. For example, for images in 3D ranging applications, the target scene corresponding to the display image is a 3D ranging scene. For example, in a face touch screen focusing mode in a photographing application, the image displayed includes a face, and the target scene is a face focusing scene. For example, in a game application, the target scene of a road-seeking image is a character road-seeking scene. In this case, the corresponding scene is directly determined as the target scene. Optionally, the scene corresponding to the display image can be obtained from the configuration information of the application program in which the display image is located.

[0115] In some other cases, the corresponding scene of the display image cannot be known, for example, in the case that the corresponding relationship between the display image and the scene is not included in the configuration information of the application program in which the display image is located. At this time, some preset common scenes can be determined as the target scene.

[0116] S340, obtaining a target feature extraction algorithm corresponding to the target scene.

[0117] It can be understood that the scene is different, and the user's point of interest or focus point is different, i.e., the meaningful point is different. For example, for a face focusing scene, the user may pay more attention to the face region. Therefore, in this case, the feature extraction algorithm commonly used for face recognition can be used as the target feature extraction algorithm. For example, for a 3D ranging scene, the user may pay more attention to a color road. Therefore, the feature extraction algorithm commonly used for color extraction can be used as the target feature extraction algorithm.

[0118] S350, performing feature extraction on each feature extraction sub-region according to the target feature extraction algorithm to obtain a feature vector corresponding to each feature extraction sub-region.

[0119] Therefore, after performing feature extraction on each feature extraction sub-region by using the target algorithm, a feature vector representing the degree of interest of the user can be extracted from each feature extraction sub-region.

[0120] According to the foregoing, in some embodiments, the target feature extraction algorithm can be one. In this case, the target feature extraction algorithm is directly used to perform feature extraction on each feature extraction sub-region to obtain a feature vector corresponding to each feature extraction sub-region.

[0121] In some other embodiments, the target scene can include multiple preset scenes, and thus the target feature extraction algorithm is multiple. In this case, referring to Figure 5 , in step S350, performing feature extraction on each feature extraction sub-region according to the target feature extraction algorithm to obtain a feature vector corresponding to each feature extraction sub-region can include step S351: performing feature extraction on the candidate feature extraction sub-region according to the target feature extraction algorithm corresponding to each preset scene to obtain a sub-feature vector of the candidate feature extraction sub-region under each target feature extraction algorithm.

[0122] The candidate feature extraction sub-region is any one of the multiple feature extraction sub-regions.

[0123] In the embodiments of the present disclosure, since the corresponding scene of the display image cannot be accurately known, the feature extraction can be performed on each feature extraction sub-region using a plurality of target feature extraction algorithms corresponding to a plurality of preset scenes respectively. That is, for each feature extraction sub-region, the feature extraction can be performed on the candidate feature extraction sub-region according to the target feature extraction algorithm corresponding to each preset scene respectively, to obtain the sub-feature vector of the candidate feature extraction sub-region under each target feature extraction algorithm, thereby obtaining the sub-feature vector of each feature extraction sub-region under each target feature extraction algorithm. That is, for any one of the feature extraction sub-regions (i.e., the candidate feature extraction sub-region), the feature extraction can be performed using the target feature extraction algorithm corresponding to each preset scene respectively, to obtain the sub-feature vector of the candidate feature extraction sub-region under each target feature extraction algorithm.

[0124] In S360, the target position point is determined based on the distance weight corresponding to each feature extraction sub-region and the feature vector corresponding to each feature extraction sub-region.

[0125] By using the above method, the feature vector extracted by the target feature extraction algorithm corresponding to the target scene corresponding to the display image can be more in line with the actual needs under the target scene, and thus the target position point determined by combining the distance weight can better meet the actual needs of the user.

[0126] Similarly, as known from the foregoing, in some embodiments, the target scene can include a plurality of preset scenes, and in this case, the feature extraction is performed on the candidate feature extraction sub-region according to the target feature extraction algorithm corresponding to each preset scene respectively, to obtain the sub-feature vector of the candidate feature extraction sub-region under each target feature extraction algorithm. For details, please continue to refer to Figure 5 , S360 can include the following steps:

[0127] In S361, the sub-feature vector of the candidate feature extraction sub-region under each target feature extraction algorithm is preprocessed to obtain the preprocessed sub-feature value of the candidate feature extraction sub-region under each target feature extraction algorithm.

[0128] Similarly, first, the sub-feature vector of the candidate feature extraction sub-region under each target feature extraction algorithm is preprocessed to obtain the numerical data, i.e., the preprocessed sub-feature value of the candidate feature extraction sub-region under each target feature extraction algorithm.

[0129] S362, obtain sub-feature values of the candidate feature extraction sub-region under each target feature extraction algorithm, the sub-feature value under a target feature extraction algorithm being a product of a distance weight corresponding to the candidate feature extraction sub-region and a pre-processed sub-feature value under the target feature extraction algorithm.

[0130] S363, perform weighted sum processing on the sub-feature values of the candidate feature extraction sub-region under each target feature extraction algorithm according to the weight values corresponding to the target feature extraction algorithms, to obtain a comprehensive feature value corresponding to the candidate feature extraction sub-region.

[0131] The weight value corresponding to each feature extraction algorithm represents the confidence of each target feature extraction algorithm under the corresponding preset scene.

[0132] It can be understood that there are various feature extraction algorithms. For example, the convolution parameters in the convolution algorithms can be different. Even if the two convolution algorithms are both used for face recognition, the convolution parameters are different, and the recognition ability for faces is different. Therefore, in the embodiments of the present disclosure, a corresponding weight value can be set for each feature extraction algorithm to represent the confidence of the target feature extraction algorithm under the corresponding preset scene.

[0133] In addition, it should be noted that the feature extraction algorithm can be set according to experience. It can also be set according to statistical values. For example, it is statistically obtained that most manufacturers or research institutes use A feature extraction algorithm for face feature extraction. Therefore, the A feature extraction algorithm can be selected as the target feature extraction algorithm under the face recognition scene.

[0134] Exemplarily, please refer to Figure 6 , which shows a schematic diagram of a determination process of a comprehensive feature value of a candidate feature extraction sub-region when a target scene includes multiple preset scenes.

[0135] As shown in Figure 6 , there are n preset scenes, the target feature extraction algorithm corresponding to scene 1 is algorithm 1, the target feature extraction algorithm corresponding to scene 2 is algorithm 2, and the target feature extraction algorithm corresponding to scene n is algorithm n. In addition, the weight corresponding to algorithm 1 is w1, the weight corresponding to algorithm 2 is w2, and the weight corresponding to algorithm n is wn.

[0136] Then, the candidate feature extraction sub-region is subjected to feature extraction by using algorithm 1, and then the feature extraction result is subjected to L2 norm processing. Then, the result after the L2 norm processing is multiplied by the weight w1, to obtain a sub-feature value 1 of the candidate feature extraction sub-region under algorithm 1.

[0137] Similarly, feature extraction is performed on the candidate feature extraction sub-region by using algorithm 1, and then the feature extraction result is subjected to L2 norm processing, and then the result after L2 norm processing is multiplied by the weight w2, to obtain the sub-feature value 2 corresponding to the candidate feature extraction sub-region under algorithm 2.

[0138] By analogy, the sub-feature value n corresponding to the candidate feature extraction sub-region under algorithm n can be obtained.

[0139] Finally, the sub-feature value 1, the sub-feature value 2, and the sub-feature value n are added, that is, the comprehensive feature value is obtained.

[0140] S364, based on the comprehensive feature value corresponding to each candidate feature extraction sub-region, the target position point is determined.

[0141] In this embodiment, the way of determining the target position point based on the feature value corresponding to each candidate feature extraction sub-region can refer to the foregoing examples. For example, when the distance weight is set to be inversely related to the ratio of the first distance to the second distance, the center point of the feature extraction sub-region with the maximum comprehensive feature value can be determined as the target position point.

[0142] By using the above method, by comprehensively considering multiple preset scenes, the position point determination with more image significance can be obtained in the case that the scene corresponding to the display image cannot be known, so that the target position point obtained is more in line with the needs of users.

[0143] Figure 7 is a structural block diagram of a position point determination apparatus 400 according to an example embodiment. Referring to Figure 7 The apparatus includes a region determination module 410, a distance weight determination module 420, a feature extraction module 430, and a target position point determination module 440. Wherein:

[0144] The region determination module 410 is configured to obtain a display region corresponding to a display image and a touch region, the touch region being obtained in response to a touch operation on the display image;

[0145] The distance weight determination module 420 is configured to obtain, based on the touch region, distance weights corresponding to each feature extraction sub-region included in the display region, the distance weight representing a distance relationship between each feature extraction sub-region and a center point of the touch region;

[0146] The feature extraction module 430 is configured to perform feature extraction on each feature extraction sub-region to obtain a feature vector corresponding to each feature extraction sub-region;

[0147] The target position point determination module 440 is configured to determine the target position point based on the distance weight corresponding to each feature extraction sub-region and the feature vector corresponding to each feature extraction sub-region.

[0148] Optionally, the distance weight determination module 420 comprises:

[0149] The distance determination sub-module is configured to obtain a first distance between the touch region center point and a candidate feature extraction sub-region center point, and obtain a second distance between the touch region center point and a farthest boundary of the touch region, the candidate feature extraction sub-region being any one of the feature extraction sub-regions;

[0150] The distance weight determination sub-module is configured to determine the distance weight corresponding to the candidate feature extraction sub-region based on the first distance and the second distance.

[0151] Optionally, the distance weight determination sub-module is further configured to obtain a ratio of the first distance and the second distance; and determine the distance weight corresponding to the candidate feature extraction sub-region according to the ratio, the distance weight being inversely related to the ratio. In this case, the target position point determination module is further configured to pre-process the feature vector corresponding to each feature extraction sub-region to obtain a pre-processed feature value corresponding to each feature extraction sub-region; obtain a product of the pre-processed feature value corresponding to each feature extraction sub-region and the distance weight corresponding to each feature extraction sub-region; and determine the center point of the feature extraction sub-region with the largest product as the target position point.

[0152] Optionally, the apparatus 400 comprises:

[0153] The feature extraction sub-region determination module is configured to determine each feature extraction sub-region around the touch region center point in the touch region with the target size as the feature extraction window.

[0154] Optionally, the feature extraction module 430 comprises:

[0155] The first obtaining sub-module is configured to obtain a target scene corresponding to the display image.

[0156] The second obtaining sub-module is configured to obtain a target feature extraction algorithm corresponding to the target scene.

[0157] The feature extraction sub-module is configured to perform feature extraction on each feature extraction sub-region according to the target feature extraction algorithm.

[0158] Optionally, the target scene includes a plurality of preset scenes, and the feature extraction submodule is further configured to perform feature extraction on the candidate feature extraction sub-region according to a target feature extraction algorithm corresponding to each preset scene, to obtain a sub-feature vector of the candidate feature extraction sub-region under each target feature extraction algorithm. In this case, the target position point determination module is further configured to pre-process the sub-feature vector of the candidate feature extraction sub-region under each target feature extraction algorithm, to obtain a pre-processed sub-feature value of the candidate feature extraction sub-region under each target feature extraction algorithm; obtain a sub-feature value of the candidate feature extraction sub-region under each target feature extraction algorithm, the sub-feature value under a target feature extraction algorithm being a product of a distance weight corresponding to the candidate feature extraction sub-region and the pre-processed sub-feature value under the target feature extraction algorithm; perform weighted sum processing on the sub-feature values of the candidate feature extraction sub-region under each target feature extraction algorithm according to a weight value corresponding to each target feature extraction algorithm, to obtain a comprehensive feature value corresponding to the candidate feature extraction sub-region, the weight value corresponding to each target feature extraction algorithm representing a confidence of each target feature extraction algorithm under the corresponding preset scene; and determine the target position point based on the comprehensive feature values corresponding to each candidate feature extraction sub-region.

[0159] As to the apparatus in the above-mentioned embodiments, the specific manners in which the modules perform operations have been described in detail in the embodiments of the method, and thus will not be described in detail here.

[0160] The present disclosure also provides a computer-readable storage medium having stored thereon computer program instructions, which, when executed by a processor, implement the steps of the position point determination method provided by the present disclosure.

[0161] Figure 8 is a block diagram of an electronic device 500 according to an example embodiment. The electronic device 500 can be, for example, a mobile phone, a notebook, a tablet computer, a desktop computer, a smart television, a wearable device, a smart robot, a vehicle terminal, etc.

[0162] Referring to Figure 8 , the electronic device 500 can include one or more of the following components: a processing component 502, a memory 504, a power component 506, a multimedia component 508, an audio component 510, an input / output (I / O) interface 512, a sensor component 514, and a communication component 516.

[0163] The processing component 502 generally controls the overall operations of the electronic device 500, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 502 can include one or more processors 520 to execute instructions to complete the above-described steps of the location point determination method, in whole or in part. Moreover, the processing component 502 can include one or more modules to facilitate the interaction between the processing component 502 and other components. For example, the processing component 502 can include a multimedia module to facilitate the interaction between the multimedia component 508 and the processing component 502.

[0164] The memory 504 is configured to store various types of data to support operations of the electronic device 500. Examples of these data include instructions to operate any applications or methods on the electronic device 500, contact data, phonebook data, messages, pictures, videos, and so on. The memory 504 can be implemented by any type of volatile or nonvolatile memory, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disc, or optical disc.

[0165] The power component 506 provides power to various components of the electronic device 500. The power component 506 can include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the electronic device 500.

[0166] The multimedia component 508 includes a screen to provide an output interface between the electronic device 500 and a user. In some embodiments, the screen can include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes the touch panel, the screen can be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense a touch, a slide, and a gesture on the touch panel. The touch sensor can not only sense a boundary of a touching or a sliding action, but also detect duration and pressure related to the touching or sliding action. In some embodiments, the multimedia component 508 includes a front camera and / or a back camera. The front camera and / or the back camera can receive external multimedia data when the electronic device 500 is in an operation mode, such as a photographing mode or a video mode. Each of the front camera and the back camera can be a fixed optical lens system or have a focal length and optical zoom capability.

[0167] The audio component 510 is configured to output and / or input audio signals. For example, the audio component 510 includes a microphone (MIC) that is configured to receive an external audio signal when the electronic device 500 is in an operation mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal can be further stored in the memory 504 or transmitted via the communication component 516. In some embodiments, the audio component 510 includes a speaker for outputting audio signals.

[0168] The I / O interface 512 provides an interface between the processing component 502 and peripheral interface modules, which can be a keypad, a click wheel, buttons, and the like. The buttons can include, but are not limited to, a home button, a volume button, a start button, and a lock button.

[0169] The sensor component 514 includes one or more sensors for providing status assessments of various aspects of the electronic device 500. For example, the sensor component 514 can detect an open / closed position of the electronic device 500, relative positioning of components, such as a display and a keypad of the electronic device 500, a change of position of the electronic device 500 or a component of the electronic device 500, presence or absence of user contact with the electronic device 500, orientation or acceleration / deceleration of the electronic device 500, and a temperature change of the electronic device 500. The sensor component 514 can include an orientation sensor, an acceleration sensor, a proximity sensor, a gesture sensor, a biometric sensor, a temperature / humidity sensor, an illumination sensor, and / or an interaction sensor.

[0170] The communication component 516 is configured to facilitate wired or wireless communication between the electronic device 500 and other devices. The electronic device 500 can access a wireless network based on a corresponding communication standard, such as WiFi, 2G, or 3G, or a combination thereof. In an example embodiment, the communication component 516 receives a broadcast signal or broadcast-related information from an external broadcasting management system via a broadcasting channel. In an example embodiment, the communication component 516 includes a Near Field Communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technology.

[0171] In an exemplary embodiment, the electronic device 500 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described location point determination method.

[0172] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 504 including instructions, which can be executed by a processor 520 of an electronic device 500 to complete the aforementioned location point determination method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0173] In another exemplary embodiment, a computer program product is also provided, which includes a computer program executable by a programmable device, the computer program having a code portion for performing the above-described location point determination method when executed by the programmable device.

[0174] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of this disclosure. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0175] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method of position point determination, characterized by The method comprises: obtaining a display region corresponding to a display image and a touch region, the touch region being obtained in response to a touch operation on the display image; obtaining a first distance between a center point of the touch region and a center point of a candidate feature extraction sub-region, and a second distance between the center point of the touch region and a farthest boundary of the touch region, the candidate feature extraction sub-region being any one of a plurality of feature extraction sub-regions, obtaining a ratio of the first distance to the second distance, determining a distance weight corresponding to the candidate feature extraction sub-region according to the ratio, the distance weight being inversely related to the ratio, the distance weight representing a distance relationship between the plurality of feature extraction sub-regions and the center point of the touch region; performing feature extraction on the plurality of feature extraction sub-regions to obtain a feature vector corresponding to each feature extraction sub-region; performing preprocessing on the feature vector corresponding to each feature extraction sub-region to obtain a preprocessed feature value corresponding to each feature extraction sub-region, obtaining a product of the preprocessed feature value corresponding to each feature extraction sub-region and the distance weight corresponding to each feature extraction sub-region, and determining a center point of a feature extraction sub-region with the largest product as a target position point.

2. The method of claim 1, wherein, The method comprises: in the touch region, determining the plurality of feature extraction sub-regions around the center point of the touch region with a target size as a feature extraction window.

3. The method according to any one of claims 1-2, characterized in that, The feature extraction on the plurality of feature extraction sub-regions comprises: obtaining a target scene corresponding to the display image; obtaining a target feature extraction algorithm corresponding to the target scene; performing feature extraction on the plurality of feature extraction sub-regions according to the target feature extraction algorithm.

4. The method of claim 3, wherein, The target scene comprises a plurality of preset scenes, and the feature extraction on the plurality of feature extraction sub-regions according to the target feature extraction algorithm comprises: performing feature extraction on a candidate feature extraction sub-region according to a target feature extraction algorithm corresponding to each preset scene to obtain a sub-feature vector corresponding to the candidate feature extraction sub-region under each target feature extraction algorithm, the candidate feature extraction sub-region being any one of the plurality of feature extraction sub-regions; The determination of the target position point based on the distance weight corresponding to each feature extraction sub-region and the feature vector corresponding to each feature extraction sub-region comprises: performing preprocessing on the sub-feature vector corresponding to the candidate feature extraction sub-region under each target feature extraction algorithm to obtain a preprocessed sub-feature value corresponding to the candidate feature extraction sub-region under each target feature extraction algorithm; obtaining a sub-feature value corresponding to the candidate feature extraction sub-region under each target feature extraction algorithm, the sub-feature value under one target feature extraction algorithm being a product of the distance weight corresponding to the candidate feature extraction sub-region and the preprocessed sub-feature value under the target feature extraction algorithm. According to the weight value corresponding to each target feature extraction algorithm, sub-feature values of the candidate feature extraction sub-region corresponding to each target feature extraction algorithm are weighted and summed to obtain a comprehensive feature value corresponding to the candidate feature extraction sub-region, and the weight value corresponding to each target feature extraction algorithm represents the confidence of each target feature extraction algorithm in the corresponding preset scene. The target position point is determined based on the comprehensive feature value corresponding to each candidate feature extraction sub-region.

5. A position point determining apparatus, characterized by comprising: The method comprises the following steps: The region determination module is configured to obtain a display region corresponding to a display image and a touch region, wherein the touch region is obtained in response to a touch operation on the display image; The distance weight determination module is configured to obtain a first distance between a center point of the touch region and a center point of a candidate feature extraction sub-region, obtain a second distance between the center point of the touch region and a farthest boundary of the touch region, obtain a ratio of the first distance to the second distance, and determine a distance weight corresponding to the candidate feature extraction sub-region according to the ratio, wherein the distance weight is inversely related to the ratio, and the distance weight represents a distance relationship between the candidate feature extraction sub-region and the center point of the touch region; The feature extraction module is configured to perform feature extraction on each feature extraction sub-region to obtain a feature vector corresponding to each feature extraction sub-region; The target position point determination module is configured to perform preprocessing on the feature vector corresponding to each feature extraction sub-region to obtain a preprocessed feature value corresponding to each feature extraction sub-region, obtain a product of the preprocessed feature value corresponding to each feature extraction sub-region and the distance weight corresponding to each feature extraction sub-region, and determine a center point of a feature extraction sub-region with the largest product as the target position point.

6. The apparatus of claim 5, wherein, The distance weight determination module comprises: The distance determination submodule is configured to obtain a first distance between a center point of the touch region and a center point of a candidate feature extraction sub-region, and obtain a second distance between the center point of the touch region and a farthest boundary of the touch region, wherein the candidate feature extraction sub-region is any one of the feature extraction sub-regions; The distance weight determination submodule is configured to determine a distance weight corresponding to the candidate feature extraction sub-region based on the first distance and the second distance.

7. An electronic device, comprising: The method comprises the following steps: A memory having a computer program stored thereon; A processor configured to execute the computer program in the memory to implement the steps of the method of any one of claims 1-4.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that The program is executed by the processor to implement the steps of the method of any one of claims 1-4.

Images