TWI772040B - Object depth information acquistition method, device, computer device and storage media - Google Patents

Abstract

The present application provides an object depth information acquisition method, an device, a computer device and a storage medium. The method includes: obtaining a plurality of images of a testing object under different characteristic light; dividing each image into a plurality of regions, and extracting luminance information of each corresponding region in the plurality of images; representing the luminance information of the same region in all images under different characteristic lights by a light source characteristic matrix; obtaining feature information of each light source feature matrix by inputting the light source feature matrix into a preset mathematical model, and grouping multiple feature information based on the similarity of feature information; setting a corresponding depth value for each group of feature information, and forming the depth information of the testing object according to the depth values corresponding to multiple groups of feature information.

Description

Object depth information acquisition method, device, computer device and storage medium

The present invention relates to the field of computer technology, in particular to a method, device, computer device and storage medium for acquiring depth information of an object.

At present, most of the methods for obtaining depth information of objects use depth cameras or dual-camera cameras. The methods for obtaining depth information of objects with depth cameras or dual cameras are expensive. In some manufacturing fields, there is an urgent need for a low-cost method to obtain objects to be measured. in-depth information.

In view of the above, it is necessary to provide a method, device, computer device and storage medium for acquiring depth information of an object, so that the acquisition method of depth information of an object can be carried out in a more economical manner.

A first aspect of the present application provides a method for acquiring depth information of an object, the method comprising: acquiring a plurality of images of an object to be measured under different characteristic lights collected by a camera device, and dividing each of the images into a plurality of regions , respectively extract the brightness information of each corresponding area in the multiple images, wherein the brightness information of the same area in all the images under different characteristic lights is represented by a light source characteristic matrix; The feature information of each of the light source feature matrices is obtained from the set mathematical model, wherein the feature information includes the gradient change of the light source feature matrix in any direction; the plurality of feature information are grouped based on the similarity of feature information, Obtain multiple sets of feature information, set a corresponding depth value for each set of feature information, and set a corresponding depth value for each set of feature information. The depth values corresponding to the group feature information constitute the depth information of the object to be measured.

Optionally, the different characteristic lights include combined characteristic lights with different luminous intensities and different luminous modes, wherein the luminous modes include one or more of: luminous angle, luminous position, spectrum, and light field distribution.

Optionally, the camera device is a single-camera camera device.

Optionally, the method includes: dividing the plurality of images according to the same preset rule.

Optionally, the preset mathematical model is a mathematical model based on a classification algorithm, wherein the classification algorithm includes any one of a binary classification algorithm, a decision tree algorithm, and a least binary algorithm.

Optionally, the training method of the preset mathematical model includes: acquiring a plurality of light source feature matrices with known feature information, dividing the light source feature matrix into a training set and a verification set; establishing a mathematical model based on a classification algorithm. , and use the training set to train the parameters of the mathematical model based on the classification algorithm; use the verification set to verify the mathematical model based on the classification algorithm after training, and use the verification set to verify the light source feature matrix in the set. Input into the mathematical model based on the classification algorithm, compare the feature information output by the model with the known feature information, and statistically obtain the prediction accuracy of the mathematical model based on the classification algorithm according to the comparison result; judging whether the prediction accuracy of the mathematical model based on the classification algorithm is less than a preset threshold; if the prediction accuracy of the mathematical model based on the classification algorithm is not less than the preset threshold, the The mathematical model of the algorithm is used as the preset mathematical model; if the prediction accuracy of the mathematical model based on the classification algorithm is less than the preset threshold, then adjust the parameters of the mathematical model based on the classification algorithm and/or Adjust the number of training set samples, and use the adjusted training set to retrain the adjusted mathematical model based on the classification algorithm, until the model prediction accuracy obtained through the verification set verification is not less than the preset threshold.

Optionally, the method for grouping the multiple sets of feature information based on the similarity of the feature information, and obtaining the multiple sets of feature information includes: matching the multiple sets of feature information according to the similarity matching method. Perform grouping, wherein the method for matching the similarity includes dividing the variation of the gradient change into a plurality of intervals according to different thresholds according to the gradient change of the feature information, and dividing the feature information in the same interval into a group .

A second aspect of the present application provides an object depth information acquisition device, the device includes: an acquisition module: used to acquire a plurality of images of the object to be measured located under different characteristic lights collected by a camera device, and each of the images is The image is divided into multiple regions, and the brightness information of each corresponding region in the multiple images is extracted respectively, wherein the brightness information of the same region in all images under different characteristic lights is represented by a light source feature matrix; the extraction module, It is used to sequentially input the light source characteristic matrix into a preset mathematical model to obtain characteristic information of each light source characteristic matrix, wherein the characteristic information includes the gradient change of the light source characteristic matrix in any direction; Group: used to group the plurality of feature information based on the similarity of the feature information to obtain multiple sets of feature information, set a corresponding depth value for each set of feature information, and form a group according to the depth values corresponding to the multiple sets of feature information. Describe the depth information of the object to be measured.

A third aspect of the present application provides a computer device, the computer device includes a processor, and the processor is configured to implement the method for acquiring depth information of an object when executing a computer program stored in a memory.

A fourth aspect of the present application provides a computer-readable storage medium, where the computer-readable storage medium stores at least one computer program, and when the at least one computer program is executed by a processor, implements the method for acquiring depth information of an object.

The object depth information acquisition method, device, computer device and storage medium of the present application can collect an image of an object to be measured through a single camera, and acquire depth information corresponding to different areas in the image through the object depth information acquisition method, so that The acquisition of object depth information is carried out in a more economical way.

1: Computer device

2: Camera device

3: Adjustable light source

10: Object depth information acquisition device

101: Get Mods

102: Extract the module

103: Execute the module

20: Storage

30: Processor

40: Computer Programs

S1~S3: Steps

In order to more clearly illustrate the technical solutions in the embodiments of the present application or in the prior art, the following The accompanying drawings required in the description of the embodiments or the prior art will be briefly introduced. Obviously, the accompanying drawings in the following description are only the embodiments of the present application. On the premise of creative work, other drawings can also be obtained based on the provided drawings.

FIG. 1 is a schematic diagram of an application environment architecture of a method for acquiring depth information of an object provided by an embodiment of the present application.

FIG. 2 is a flowchart of a method for acquiring depth information of an object provided by an embodiment of the present application.

FIG. 3 is a schematic structural diagram of an apparatus for acquiring depth information of an object provided by an embodiment of the present application.

FIG. 4 is a schematic diagram of a computer device provided by an embodiment of the present application.

In order to more clearly understand the above objects, features and advantages of the present application, the present application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments may be combined with each other in the case of no conflict.

In the following description, many specific details are set forth to facilitate a full understanding of the present application, and the described embodiments are only a part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this application belongs. The terms used herein in the specification of the application are for the purpose of describing specific embodiments only, and are not intended to limit the application.

Referring to FIG. 1 , it is a schematic diagram of an application environment architecture of the method for acquiring depth information of an object provided by an embodiment of the present application.

The object depth information acquisition method in this embodiment is applied in the computer device 1, and the The computer device 1 establishes a communication connection with the camera device 2 and the adjustable light source 3 through a network. The network may be a wired network or a wireless network, such as radio, Wireless Fidelity (WIFI), cellular, satellite, broadcast, and the like. The adjustable light source 3 is used to emit light with different characteristics, and the light with different characteristics refers to the combined characteristic light of different luminous intensities and different luminous modes, wherein the luminous modes include: luminous angle, luminous position, spectrum , one or more of the light field distributions. The camera device 2 is used for collecting image information of an object under different characteristic light irradiation, and sending the image information to the computer device 1 . The computer device 1 is used to switch different luminous intensities and luminous modes of the adjustable light source 3 to emit light with the different characteristics, and is also used to receive the image information collected by the camera device 2 to obtain the brightness information of the image information. , and extract the feature information of the luminance information, group the feature information according to the preset method, and set the corresponding depth value for each group of feature information according to the preset rule, Combined to obtain object depth information.

The computer device 1 may be an electronic device installed with object depth information acquisition software, such as a personal computer, a server, and the like, wherein the server may be a single server, a server cluster, or a cloud server.

The camera device 2 is a single-camera camera device with a shooting function, including but not limited to a mobile phone, a camera, a tablet computer, a monitor, and the like.

One or more of the light-emitting intensity, light-emitting angle, light-emitting position, spectrum, and light-field distribution of the adjustable light source 3 can be adjusted, thereby realizing light-emitting in different modes. In one embodiment, the above-mentioned adjustable light sources 3 may include, but are not limited to, adjustable LED lights, adjustable light boxes, and the like.

In one embodiment, the computer device 1 may be located in the camera device 2, and the computer device is configured to receive image information collected by the camera device 2, and acquire depth information of the image according to a method for acquiring depth information of an object.

Please refer to FIG. 2 , which is a flowchart of a method for acquiring depth information of an object provided by an embodiment of the present application. According to different requirements, the order of the steps in the flowchart can be changed, and some steps can be omitted.

Step S1: Acquire multiple images of the object to be tested under different characteristic lights collected by the camera, divide each of the images into multiple regions, and extract the brightness of each corresponding region in the multiple images respectively. information, in which the luminance information of the same area in all images under different characteristic light is represented by a light source characteristic matrix.

The specific implementation steps of step S1 are as follows:

(1) Acquiring multiple images of the object to be measured under different characteristic lights and collected by the camera device.

The different characteristic lights refer to the combined characteristic lights of different luminous intensities and different luminous modes, wherein the luminous modes include one or more of: luminous angle, luminous position, spectrum, and light field distribution.

In the same light-emitting mode, the light-emitting intensity of the light source is set to M levels, and for each level of light-emitting intensity, the camera 2 acquires an image of the object. For the same light-emitting mode and light sources with different light-emitting intensities, a total of M images are obtained. If the light-emitting modes include N types, the total number of acquired images is M×N. For example, the luminous intensity of the light source is divided into 100 levels. For each level of luminous luminous intensity, images of objects in different luminous modes are obtained, that is, the images of objects in 4 luminous modes: luminous angle, luminous position, spectrum, and light field distribution. The total number of images captured by the object under the light source with 100-level luminous intensity and 4 light-emitting modes is 100×4 images.

(2) The computer device 1 divides each of the images into a plurality of regions.

In one embodiment, the plurality of images are divided in the same manner. The division method may be to divide the horizontal axis direction and the vertical axis direction of the image in equal or preset proportions, and divide the image into H×V regions, where H and V are positive. Integer, H represents the number of segments divided in the horizontal axis direction of the image, and V represents the number of segments divided in the vertical axis direction of the image. The dividing manner may further be that according to an edge of the image as a reference, the image is divided into a plurality of regions with an oblique line at a preset angle with the reference edge.

(3) The computer device 1 separately extracts the luminance information of each corresponding region in the plurality of images.

The method for recalling the image brightness includes converting the RGB values of the primitive points in the image through a color gamut conversion method to convert the RGB values of the primitive points into Yuv values, wherein Y represents the RGB value of the primitive points. Luminance information, uv represents the chromaticity information of the primitive point, and the color gamut conversion method includes any of a neural network-based color gamut conversion method, a tetrahedral-based color gamut conversion method, and a linear regression-based color gamut conversion method. A sort of.

(4) The computer device 1 represents the luminance information of the same area in all images under different characteristic lights by a light source characteristic matrix.

The size of the matrix is M×N, where M×N represents the number of different characteristic lights. For example, the computer device 1 divides the image collected by the camera device 2 into H×V areas according to a preset method, and for any area, obtains the brightness information of the area under the illumination of M×N characteristic light sources, and uses the acquired The M×N pieces of luminance information form a light source characteristic matrix, and the element a01 in the matrix represents the luminance information of the region when the luminous intensity of the light source is 0 and the luminous mode is the first mode. By analogy, the brightness information of other regions in the image under different characteristic light sources is obtained, and multiple sets of brightness information of different regions are used to form multiple light source characteristic matrices, wherein the number of the light source characteristic matrices is related to the object. The number of image segmentation regions is equal.

Step S2 , sequentially inputting the light source characteristic matrix into a preset mathematical model to obtain characteristic information of each light source characteristic matrix, wherein the characteristic information includes the gradient change of the light source characteristic matrix in any direction.

Inputting the light source characteristic matrix into a preset mathematical model in turn, extracting characteristic information of the light source characteristic matrix, the number of the characteristic information is equal to the number of the light source characteristic matrix, in one embodiment, to be tested The image of the object is divided into H×V regions, and the H×V regions correspond to H×V optical feature matrices. The H×V optical feature matrices are input into the preset data model, and the extracted feature information The number is H×V. The feature information includes the gradient change of the luminance information in any direction, and the preset mathematical model obtains the feature information in the light source feature matrix through a mathematical model based on a classification algorithm, and the classification algorithm includes but Not limited to based on binary classification algorithm, decision tree algorithm, least binary algorithm.

In one embodiment, the preset mathematical model is a mathematical model based on a classification algorithm, and the training method of the preset mathematical model includes: acquiring a plurality of light source feature matrices with known feature information; The feature matrix is divided into a training set and a verification set; a mathematical model based on a classification algorithm is established, and the training set is used to train the parameters of the mathematical model based on the classification algorithm; The mathematical model based on the classification algorithm is used for verification, the light source feature matrix in the verification set is input into the mathematical model based on the classification algorithm, the feature information output by the model is compared with the known feature information, and according to the comparison Statistically obtain the prediction accuracy of the mathematical model based on the classification algorithm; determine whether the prediction accuracy of the mathematical model based on the classification algorithm is less than a preset threshold; if the prediction accuracy of the mathematical model based on the classification algorithm is accurate If the accuracy rate is not less than the preset threshold, the trained mathematical model based on the classification algorithm is used as the preset mathematical model; if the prediction accuracy of the mathematical model based on the classification algorithm is less than the preset mathematical model threshold, then adjust the parameters of the mathematical model based on the classification algorithm and/or adjust the number of training set samples, and use the adjusted training set to retrain the adjusted mathematical model based on the classification algorithm until The model prediction accuracy obtained through the verification set is not less than the preset threshold.

Step S3, grouping the plurality of feature information based on the similarity of the feature information to obtain multiple sets of feature information, setting a corresponding depth value for each set of feature information, and forming the Depth information of the object to be measured.

In one embodiment, the plurality of, for example, H×V pieces of feature information mentioned in step S2 are grouped according to a similarity matching method, wherein the similarity matching method includes: combining the plurality of features The information is sorted according to the preset rules; the sorted feature information is sorted according to the preset rules The interval is divided into different intervals, at least one feature information in the same interval is divided into a group, and corresponding depth values are assigned to the feature information in the same group, and the depth information of the object to be measured is composed of multiple groups of depth values.

In one embodiment, the H×V feature information is grouped using a classification method based on a machine learning algorithm. The machine learning algorithm includes the classification method of the support vector machine algorithm, the Bayesian classification algorithm, and the like. The H×V features are assigned corresponding depth values to the feature information in the same group through the machine learning algorithm, and the depth information of the object to be measured is composed of a plurality of sets of depth values.

The above-mentioned Fig. 2 describes the object depth information acquisition method of the present application in detail. Below, in conjunction with Figs. 3-4, the functional modules of the software device for realizing the object depth information acquisition method and the hardware for realizing the object depth information acquisition method are described in detail. The device architecture is introduced.

It should be understood that the embodiments are only used for illustration, and are not limited by this structure in the scope of the patent application.

FIG. 3 is a structural diagram of a preferred embodiment of an object depth information acquisition device according to the present application.

In some embodiments, the object depth information acquisition device 10 is executed in a computer device. The computer device is connected to a plurality of user terminals through a network. The object depth information acquisition device 10 may include a plurality of functional modules composed of program code segments. The code of each program segment in the object depth information acquisition device 10 can be stored in the memory of the computer device and executed by the at least one processor to realize the object depth information acquisition function.

In this embodiment, the object depth information acquiring apparatus 10 can be divided into a plurality of functional modules according to the functions performed by the apparatus 10 . Referring to FIG. 3 , the functional modules may include: an acquisition module 101 , an extraction module 102 , and an execution module 103 . The module referred to in this application refers to a series of computer program segments that can be executed by at least one processor and can perform fixed functions, and are stored in a memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.

The acquisition module 101 is used to acquire the object to be measured collected by the camera device located in a different location. For a plurality of images under characteristic light, each image is divided into a plurality of regions, and the brightness information of each corresponding region in the plurality of images is extracted respectively, wherein the same region in all images is under different characteristic light. The luminance information below is represented by a light source characteristic matrix.

The specific implementation steps of the acquisition module 101 are as follows:

(1) The acquisition module 101 acquires a plurality of images of the object to be measured under different characteristic lights and collected by the camera device.

The different characteristic lights refer to the combined characteristic lights of different luminous intensities and different luminous modes, wherein the luminous modes include one or more of: luminous angle, luminous position, spectrum, and light field distribution.

In the same light-emitting mode, the light-emitting intensity of the light source is set to M levels, and for each level of light-emitting intensity, the camera 2 acquires an image of the object. For the same light-emitting mode and light sources with different light-emitting intensities, a total of M images are obtained. If the light-emitting modes include N types, the total number of acquired images is M×N. For example, the luminous intensity of the light source is divided into 100 levels. For each level of luminous luminous intensity, images of objects in different luminous modes are obtained, that is, the images of objects in 4 luminous modes: luminous angle, luminous position, spectrum, and light field distribution. The total number of images captured by the object under the light source with 100-level luminous intensity and 4 light-emitting modes is 100×4 images.

(2) The acquisition module 101 divides each image into a plurality of regions.

In one embodiment, the plurality of images are divided in the same manner. The division method may be to divide the horizontal axis direction and the vertical axis direction of the image in equal or preset proportions, and divide the image into H×V regions, where H and V are positive. Integer, H represents the number of segments divided in the horizontal axis direction of the image, and V represents the number of segments divided in the vertical axis direction of the image. The dividing manner may further be that according to an edge of the image as a reference, the image is divided into a plurality of regions with an oblique line at a preset angle with the reference edge.

(3) The acquiring module 101 extracts the luminance information of each corresponding region in the plurality of images respectively.

The method for recalling the brightness of the image includes converting the RGB color of the primitive points in the image. The degree value converts the RGB value of the primitive point into the Yuv value through the color gamut conversion method, wherein Y represents the luminance information of the primitive point, and uv represents the chromaticity information of the primitive point. The color gamut conversion method includes a neural network based Any one of the color gamut conversion method based on the road, the color gamut conversion method based on the tetrahedron, and the color gamut conversion method based on the linear regression.

(4) The acquisition module 101 represents the luminance information of the same area in all images under different characteristic lights by a light source characteristic matrix.

The size of the matrix is M×N, where M×N represents the number of different characteristic lights. For example, the computer device 1 divides the image collected by the camera device 2 into H×V areas according to a preset method, and for any area, obtains the brightness information of the area under the illumination of M×N characteristic light sources, and uses the acquired The M×N pieces of luminance information form a light source characteristic matrix, and the element a ₀₁ in the matrix represents the luminance information of the region when the luminous intensity of the light source is 0 and the luminous mode is the first mode. By analogy, the brightness information of other regions in the image under different characteristic light sources is obtained, and multiple sets of brightness information of different regions are used to form multiple light source characteristic matrices, wherein the number of the light source characteristic matrices is related to the object. The number of image segmentation regions is equal.

The extraction module 102 is used to sequentially input the light source feature matrix into a preset mathematical model to obtain feature information of each light source feature matrix, wherein the feature information includes the light source feature matrix in any direction. gradient change on .

Inputting the light source characteristic matrix into a preset mathematical model in turn, extracting characteristic information of the light source characteristic matrix, the number of the characteristic information is equal to the number of the light source characteristic matrix, in one embodiment, to be tested The image of the object is divided into H×V regions, and the H×V regions correspond to H×V optical feature matrices. The H×V optical feature matrices are input into the preset data model, and the extracted feature information The number is H×V. The feature information includes the gradient change of the luminance information in any direction, and the preset mathematical model obtains the feature information in the light source feature matrix through a mathematical model based on a classification algorithm, and the classification algorithm includes but Not limited to based on binary classification algorithm, decision tree algorithm, least binary algorithm.

In one embodiment, the preset mathematical model is a mathematical model based on a classification algorithm The training method of the preset mathematical model includes: acquiring a plurality of light source characteristic matrices with known characteristic information, dividing the light source characteristic matrix into a training set and a verification set; establishing a mathematical model based on a classification algorithm, and Use the training set to train the parameters of the mathematical model based on the classification algorithm; use the verification set to verify the trained mathematical model based on the classification algorithm, and input the light source feature matrix in the verification set into the In the mathematical model based on the classification algorithm, the feature information output by the model is compared with the known feature information, and the prediction accuracy of the mathematical model based on the classification algorithm is statistically obtained according to the comparison result; Whether the prediction accuracy of the mathematical model based on the classification algorithm is less than a preset threshold; if the prediction accuracy of the mathematical model based on the classification algorithm is not less than the preset threshold, then the classification algorithm based on the training completed The mathematical model based on the classification algorithm is used as the preset mathematical model; if the prediction accuracy of the mathematical model based on the classification algorithm is less than the preset threshold, then adjust the parameters of the mathematical model based on the classification algorithm and/or adjust the training The number of samples in the set is determined, and the adjusted mathematical model based on the classification algorithm is retrained using the adjusted training set until the model prediction accuracy obtained through the verification set is not less than the preset threshold.

The execution module 103 is configured to group the plurality of feature information based on the similarity of the feature information to obtain multiple sets of feature information, set a corresponding depth value for each set of feature information, and correspond to the multiple sets of feature information according to the The depth value of , constitutes the depth information of the object to be measured.

In one embodiment, the plurality of, for example, H×V pieces of feature information mentioned in the extraction module 102 are grouped according to a similarity matching method, wherein the similarity matching method includes: The plurality of feature information is sorted according to a preset rule; the sorted feature information is divided into different intervals according to a preset interval, and at least one feature information in the same interval is divided into In one group, corresponding depth values are assigned to the feature information in the same group, and the depth information of the object to be measured is composed of multiple groups of depth values.

In one embodiment, the H×V feature information is grouped using a classification method based on a machine learning algorithm. The machine learning algorithm includes the classification method of the support vector machine algorithm, the Bayesian classification algorithm, and the like. The H×V features are assigned corresponding depth values to the feature information in the same group through the machine learning algorithm, and the depth information of the object to be measured is composed of a plurality of sets of depth values.

FIG. 4 is a schematic diagram of a preferred embodiment of the computer device 1 of the present application.

The computer device 1 includes a storage 20, a processor 30, and a computer program 40 stored in the storage 20 and executable on the processor 30, such as an object depth information acquisition program. When the processor 30 executes the computer program 40 , the steps in the above-mentioned embodiment of the method for acquiring depth information of an object are implemented, for example, steps S1 to S3 shown in FIG. 2 . Alternatively, when the processor 30 executes the computer program 40 , the functions of each module/unit in the above-mentioned embodiment of the apparatus for acquiring depth information of an object are realized, for example, the units 101 to 103 in FIG. 3 .

Exemplarily, the computer program 40 may be divided into one or more modules/units, and the one or more modules/units are stored in the storage 20 and executed by the processor 30 , to complete this application. The one or more modules/units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program 40 in the computer device 1 . For example, the computer program 40 can be divided into an acquisition module 101 , an extraction module 102 , and an execution module 103 in FIG. 3 .

The computer device 1 may be a computing device such as a desktop computer, a notebook, a palmtop computer, and a cloud server. Those skilled in the art can understand that the schematic diagram is only an example of the computer device 1, and does not constitute a limitation on the computer device 1. It may include more or less components than the one shown, or combine some components, or different Components, such as the computer device 1, may also include input and output devices, network access devices, bus bars, and the like.

The so-called processor 30 may be a central processing unit (Central Processing Unit, CPU), other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs) ) or other programmable logic elements, discrete gate or transistor logic elements, discrete hardware elements, etc. The general-purpose processor can be a microprocessor or the processor 30 can also be any conventional processor, etc. The processor 30 is the control center of the computer device 1, and uses various interfaces and lines to connect the entire computer device 1. various parts.

The storage 20 can be used to store the computer programs 40 and/or modules/units, and the processor 30 runs or executes the computer programs and/or modules/units stored in the storage 20, and The data stored in the storage 20 is called to realize various functions of the computer device 1 . The storage 20 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; storage data The area can store data (such as audio data, phone book, etc.) created according to the use of the computer device 1 and the like. In addition, the storage 20 may include high-speed random access storage, and may also include non-volatile storage, such as hard disk, storage, plug-in hard disk, Smart Media Card (SMC), Secure Digital (Secure Digital, SD) card, flash memory card (Flash Card), at least one disk storage element, flash memory element, or other volatile solid state storage element.

If the modules/units integrated in the computer device 1 are implemented in the form of software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium. Based on this understanding, the present application can implement all or part of the processes in the methods of the above embodiments, and can also be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, so When the computer program is executed by the processor, the steps of the above-mentioned method embodiments can be implemented. Wherein, the computer program includes computer program code, and the computer program code may be in the form of original code, object code, executable file, or some intermediate form. The computer-readable storage medium may include any computer program code capable of carrying Entity or device, recording medium, flash drive, portable hard disk, magnetic disk, optical disc, computer storage, ROM (Read-Only Memory), random access memory (RAM, Random Access Memory), electrical Carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable storage medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, computer-readable Storage media does not include electrical carrier signals and telecommunication signals.

In the several embodiments provided in this application, it should be understood that the disclosed computer apparatus and method may be implemented in other manners. For example, the computer apparatus embodiments described above are only illustrative. For example, the division of the units is only a logical function division, and other division methods may be used in actual implementation.

In addition, each functional unit in each embodiment of the present application may be integrated in the same processing unit, or each unit may exist physically alone, or two or more units may be integrated in the same unit. The above-mentioned integrated units can be implemented in the form of hardware, or can be implemented in the form of hardware plus software function modules.

It will be apparent to those skilled in the art that the present application is not limited to the details of the above-described exemplary embodiments, but that the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application. Accordingly, the embodiments are to be regarded in all respects as illustrative and not restrictive, and the scope of this application is defined by the appended claims rather than the foregoing description, and is therefore intended to fall within the scope of the claims. All changes within the meaning and scope of the equivalents of , are included in this application. Any reference sign in a claim should not be construed as limiting the claim to which it relates. Furthermore, it is clear that the word "comprising" does not exclude other units or steps and the singular does not exclude the plural. A plurality of units or computer devices stated in the computer device claim may also be implemented by the same unit or computer device through software or hardware. The terms first, second, etc. are used to denote names and do not denote any particular order.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit them. Although the present application has been described in detail with reference to the preferred Persons should understand that the technical solutions of the present application may be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present application.

S1~S3: Steps

Claims (9)

A method for acquiring depth information of an object, wherein the method comprises: acquiring a plurality of images of an object to be measured under different characteristic lights collected by a camera, dividing each of the images into a plurality of regions, and extracting the The brightness information of each corresponding area in the multiple images, wherein the brightness information of the same area in all the images under different characteristic lights is represented by a light source characteristic matrix; the light source characteristic matrix is sequentially input into the preset mathematical model Obtain feature information of each of the light source feature matrices, wherein the feature information includes the gradient change of the light source feature matrix in any direction; group the plurality of feature information based on the similarity of the feature information to obtain multiple sets of feature information , set a corresponding depth value for each set of feature information, and form the depth information of the object to be measured according to the depth values corresponding to the multiple sets of feature information, wherein the similarity based on the feature information is used for the multiple features. Information grouping, and the method for obtaining multiple sets of feature information includes: grouping the multiple sets of feature information according to a similarity matching method, wherein the similarity matching method includes changing the gradient according to the gradient change of the feature information. The change amount of the change is divided into multiple intervals according to different thresholds, and the feature information in the same interval is divided into a group. The method for acquiring depth information of an object according to claim 1, wherein the different characteristic lights comprise combined characteristic lights of different luminous intensities and different luminous modes, wherein the luminous modes include: luminous angle, luminous position, spectrum, light One or more of the field distributions. The object depth information acquisition method according to claim 1, wherein the camera device is a single-camera camera device. The method for acquiring depth information of an object according to claim 1, wherein the method comprises: dividing the plurality of images according to the same preset rule. The object depth information acquisition method according to claim 1, wherein the preset mathematical model is a mathematical model based on a classification algorithm, wherein the classification algorithm includes a binary classification algorithm, a decision tree algorithm, Any of the least binary algorithms. The method for acquiring depth information of an object according to claim 5, wherein the training method of the preset mathematical model comprises: acquiring a plurality of light source feature matrices of known feature information, and dividing the light source feature matrix into a training set and a verification set; establish a mathematical model based on the classification algorithm, and use the training set to train the parameters of the mathematical model based on the classification algorithm; use the verification set to train the mathematical model based on the classification algorithm The model is verified, the light source feature matrix in the verification set is input into the mathematical model based on the classification algorithm, the feature information output by the model is compared with the known feature information, and the statistics based on the comparison result are obtained. The prediction accuracy of the mathematical model of the classification algorithm; determine whether the prediction accuracy of the mathematical model based on the classification algorithm is less than a preset threshold; if the prediction accuracy of the mathematical model based on the classification algorithm is not less than the preset threshold, then the trained mathematical model based on the classification algorithm is used as the preset mathematical model; if the prediction accuracy of the mathematical model based on the classification algorithm is less than the preset threshold, then adjust the mathematical model based on the classification algorithm The parameters of the mathematical model of the classification algorithm and/or the number of samples in the adjusted training set, and the adjusted mathematical model based on the classification algorithm is retrained using the adjusted training set until it is verified through the verification set. The prediction accuracy of the model is not less than the preset threshold. An object depth information acquisition device, wherein the device includes: an acquisition module: used to acquire a plurality of images of an object to be measured under different characteristic lights collected by a camera device, and divide each of the images into a plurality of region, respectively extracting the brightness information of each corresponding region in the multiple images, wherein the brightness information of the same region in all images under different characteristic lights is represented by a light source feature matrix; an extraction module is used to extract the The light source characteristic matrix is sequentially input into a preset mathematical model to obtain characteristic information of each light source characteristic matrix, wherein the characteristic information includes the gradient change of the light source characteristic matrix in any direction; Execution module: used to group the plurality of feature information based on the similarity of the feature information, obtain multiple sets of feature information, set a corresponding depth value for each set of feature information, and set a corresponding depth value according to the multiple sets of feature information. The depth information of the object to be measured is composed, wherein the method of grouping the plurality of feature information based on the similarity of the feature information, and obtaining the plurality of sets of feature information includes: matching the plurality of sets of feature information according to the similarity matching method. Perform grouping, wherein the method for matching the similarity includes dividing the variation of the gradient change into a plurality of intervals according to different thresholds according to the gradient change of the feature information, and dividing the feature information in the same interval into a group . A computer device, wherein: the computer device includes a processor, and the processor is configured to implement the method for acquiring object depth information according to any one of claim 1 to 6 when executing a computer program stored in a memory. A computer-readable storage medium, wherein the computer-readable storage medium stores at least one computer program, and when the at least one computer program is executed by a processor, realizes the object depth according to any one of claim 1 to 6 Information acquisition method.

Images