A method for associating basic information of characters based on view data

Abstract

The invention discloses a basic information association method for a person based on view data, and the invention relates to the technical field of view person analysis. The basic information association method is as follows: view data of the same position is obtained by an optical sensor, an optical imaging camera and a photographic camera, and the view data is classified. According to the classification of the view data, the content in the view is analyzed, three color spaces are established, and infrared imaging person features in optical imaging are extracted. The advantage of the invention is that the infrared imaging camera and the photographic camera are used to shoot at the same position, and then the basic features of the person are extracted, so that the person feature information in the color space will be continuously improved, and then the optical sensor is used to supplement the person information captured by the photographic camera, and the person feature information and the person skin color in the color space are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of view character analysis, and in particular to a method for associating basic information of characters based on view data. Background Art

[0002] In modern society, with the continuous development and popularization of optical application technology, obtaining view data has become a common way. The view data obtained by optical equipment often contains rich information, among which the image of people is an important data source in many scenes. Using the view data of optical applications to associate the basic information of people is of great significance in various fields. It can improve the utilization value and application effect of data and promote the development and progress of related fields.

[0003] However, different light intensities may affect the camera's shooting of a person, which may affect the accuracy of the person's information association when the person is subsequently associated with the basic information. Therefore, we propose a basic information association method for people based on view data. Summary of the invention

[0004] The object of the present invention is to provide a method for associating basic information of characters based on view data.

[0005] In order to solve the problems raised in the above background technology, the present invention provides the following technical solution: a basic information association method of a character based on view data, the basic information association method is as follows:

[0006] S1. Obtaining view data of the same position through an optical sensor, an optical imaging camera, and a photographic camera, and classifying the view data;

[0007] S2. Analyze the content in the view according to the classification of the view data and establish three color spaces;

[0008] S3, extracting the character features of infrared imaging in optical imaging, and transmitting the extracted features to the color space, establishing a character framework in the color space according to the character features, and generating a corresponding character feature framework, including the positions of key points in the character features, and analyzing the character posture information to obtain subsequent corresponding character posture data;

[0009] S4, obtaining the data of S1 and S3, and analyzing the person information in the view captured by the camera, matching the position of the person feature in S3 with the position in the view, and cropping the position of the person in the view;

[0010] S5, obtaining cropped photo data of the view, and then using an optical sensor to obtain optical data in the current environment, and restoring the color of the person information in the cropped photo data by analyzing the optical data;

[0011] S6, performing color segmentation on the character information in the view, and extracting the color segmentation and the character information features to the positions of the character feature frames respectively;

[0012] S7, obtaining the person posture data of S3 and processing the person features in different views, then retrieving the person feature frame in the color space for matching, and using the evaluation network to evaluate the person information in the color space, and finally storing the person feature information with the highest evaluation value in the color space;

[0013] S8. Obtain character database information, and compare the character database information with the character information in the view.

[0014] As a further solution of the present invention: in S1, the views captured by the optical imaging and the photographic camera are classified, and the view data at different times are marked and classified. The views of the optical imaging at different times are marked as Aa, Ab...Azz, and the views captured by the photographic camera at different times are marked as Ba, Bb...Bzz. The view data at the same time are retrieved for classification, wherein the character features in the optical imaging views are respectively marked as Aa1, Aa2...Aan, and the character feature information in the photographic camera views are respectively marked as Ba1, Ba2...Ban.

[0015] As a further solution of the present invention: in S2, the color space is divided according to the classification status to obtain a view processing space, a color contrast space and a color storage space respectively. The view processing space mainly processes view data at the same time, and the color contrast space is used for the system to match and associate the character information on the view with the character information inside the color storage space after processing the character information on the view.

[0016] As a further solution of the present invention: in S3, character feature information in the view is obtained, and according to the number of characters and character features in the view, character data is transmitted to the color space respectively, and a basic framework of the character feature information is established in the view processing space.

[0017] As a further solution of the present invention: in S4, the position of the view captured by the photographic camera is cropped according to the position of the character feature in the view, ensuring that the cropped positions are the same in the same view classification, and further ensuring that the cropped positions are the same in the two views.

[0018] As a further solution of the present invention: in S5, optical data in the optical sensor is retrieved, wherein the optical data includes light intensity and spectrum, and the basic color of the character is restored according to the cropped character information and the optical data, and the color of the overexposed and too dark character information is restored.

[0019] As a further solution of the present invention: in S6, after the character's color is restored, the color of the character in the view is segmented and extracted, the character's color with a threshold value within 5 color difference values ​​is extracted, and the extracted color is supplemented into the character feature framework in the color space according to the position of the character's key points.

[0020] As a further solution of the present invention: after the character feature processing of the view is completed in S7, the character feature information stored in the color space is matched with the processed information, and then the evaluation network evaluates the two character feature information to obtain the best character feature information. When the evaluation value is greater than 90%, the color space stops updating the current character feature data.

[0021] As a further solution of the present invention: after obtaining the character database information in S8, the character database information is matched with the basic feature information of the character in the color space to obtain the basic information of the character, and the latest feature information of the character is updated to the character database.

[0022] By adopting the above technical solution, compared with the prior art, the beneficial effects of the present invention are:

[0023] 1. The present invention uses an infrared imaging camera and a photographic camera to shoot at the same position, thereby extracting basic features of a person, so that the person feature information in the color space is continuously improved, and then uses an optical sensor to supplement the person information captured by the photographic camera to improve the person feature information and the person skin color in the color space;

[0024] 2. The present invention matches the character features with the character features in the color storage space after the character features are perfected, and then matches the basic information in the retrieved character database, uses the character features to match the character information in the character database, retrieves the character information in the view, obtains the best character feature information through the evaluation network, and then transmits the best character feature information to the color storage space, thereby improving the accuracy of subsequent character feature matching;

[0025] 3. When the character feature evaluation value of the color storage space is greater than 90%, the character feature with an evaluation value greater than 90% is transmitted to the character database, and the basic feature information of the character in the character database is updated, thereby improving the character information in the character database. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 1 is a step diagram of a method for associating character information in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] The specific embodiments of the present invention are further described below in conjunction with the accompanying drawings. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0028] The present invention provides a basic information association method for characters based on view data, and the basic information association method is as follows:

[0029] S1. Obtaining view data of the same position through an optical sensor, an optical imaging camera, and a photographic camera, and classifying the view data;

[0030] S2. Analyze the content in the view according to the classification of the view data and establish three color spaces;

[0031] S3, extracting the character features of infrared imaging in optical imaging, and transmitting the extracted features to the color space, establishing a character framework in the color space according to the character features, and generating a corresponding character feature framework, including the positions of key points in the character features, and analyzing the character posture information to obtain subsequent corresponding character posture data;

[0032] S4, obtaining the data of S1 and S3, and analyzing the person information in the view captured by the camera, matching the position of the person feature in S3 with the position in the view, and cropping the position of the person in the view;

[0033] S5, obtaining cropped photo data of the view, and then using an optical sensor to obtain optical data in the current environment, and restoring the color of the person information in the cropped photo data by analyzing the optical data;

[0034] S6, performing color segmentation on the character information in the view, and extracting the color segmentation and the character information features to the positions of the character feature frames respectively;

[0035] S7, obtaining the person posture data of S3 and processing the person features in different views, then retrieving the person feature frame in the color space for matching, and using the evaluation network to evaluate the person information in the color space, and finally storing the person feature information with the highest evaluation value in the color space;

[0036] S8. Obtain character database information, and compare the character database information with the character information in the view.

[0037] In one embodiment of the present invention: S1 classifies the views captured by the optical imaging and the photographic camera, and marks and classifies the view data at different times, wherein the views of the optical imaging at different times are marked as Aa, Ab...Azz, and the views captured by the photographic camera at different times are marked as Ba, Bb...Bzz, and the view data at the same time are retrieved for classification, wherein the character features in the optical imaging views are respectively marked as Aa1, Aa2...Aan, and the character feature information in the photographic camera views are respectively marked as Ba1, Ba2...Ban.

[0038] In one embodiment of the present invention: in S2, the color space is divided according to the classification status to obtain a view processing space, a color contrast space and a color storage space respectively. The view processing space mainly processes the view data of the same time, and the color contrast space is used for the system to match and associate the character information on the view with the character information in the color storage space after processing the character information on the view.

[0039] In one embodiment of the present invention: in S3, character feature information in the view is obtained, and character data is transmitted to the color space respectively according to the number and characteristics of characters in the view, and a basic framework of the character feature information is established in the view processing space.

[0040] In one embodiment of the present invention: In S4, the position in the view captured by the camera is cropped according to the position of the character feature in the view, ensuring that the cropped position is the same in the same view classification, and further ensuring that the cropped position is the same in the two views.

[0041] In one embodiment of the present invention: in S5, optical data in the optical sensor is retrieved, wherein the optical data includes light intensity and spectrum, and the basic color of the character is restored according to the cropped character information and the optical data, and the color of the overexposed and too dark character information is restored.

[0042] In one embodiment of the present invention: in S6, after the character's color is restored, the color of the character in the view is segmented and extracted, the character color with a threshold value within 5 color difference values ​​is extracted, and the extracted color is supplemented into the character feature framework in the color space according to the position of the character's key points.

[0043] In one embodiment of the present invention: after the view character feature processing is completed in S7, the character feature information stored in the color space is matched with the processed information, and then the evaluation network evaluates the two character feature information to obtain the best character feature information. When the evaluation value is greater than 90%, the color space stops updating the current character feature data.

[0044] In one embodiment of the present invention: in S8, after acquiring the character database information, the character database information is matched with the basic feature information of the character in the color space to acquire the basic character information, and at the same time, the latest feature information of the character is updated to the character database.

[0045] Embodiment 1

[0046] The character feature framework is determined by infrared imaging conditions, and then the feature information of the key points of the character is determined. By obtaining view data of imaging by different devices at the same position, the optical sensor obtains the view data affected by the current light, and supplements the light state in the view data, and supplements the supplemented character color information to the character feature framework, thereby improving the character data captured in different light, and improving the subsequent correlation method of basic character information. The system needs to obtain the number of character features in the infrared imaging view through the formula, and the specific formula is as follows: Aa= (Aa1, Aa2...Aan), Ab= (Ab1, Ab2...Abn)...Azz= (Azz1, Azz2...Azzn), and the number of people photographed by the camera is obtained through the formula, and the specific formula is as follows: Ba=Q1%Ba1, Q2%Ba2...Qn%Ban), where Q1%, Q2%...Qn% all represent the proportion of people affected by light in the view.

[0047] Embodiment 2

[0048] The character feature information is processed through the view processing space, color contrast space and color storage space respectively, that is, after the views of Aa and Ba are processed in the view processing space, the view data in the view processing space is transferred to the color contrast space, and the character feature information in the color contrast space is compared with the character feature information in the color storage space to obtain character information at different angles and more complete features in the character information. Then, the character feature information value is established through the evaluation network, and the character information with a higher feature information value is placed in the color storage space. The character feature information is established by the color storage space, and the character evaluation value in the color storage space is established through the character key points and the degree of completeness of the character features. The formula is used to determine whether the character data in the color storage space needs to be updated. The specific formula is as follows: M=((Aa1, Ba1)X1%) / ((Ab1, Bb1)X2%), when M﹤1, the character data in the color storage space is updated, and when M﹥1, the character data in the color storage space is not updated, where X1% and X2% respectively represent the proportion of the character features displayed in the current view.

[0049] Embodiment 3

[0050] The character features are matched with the information in the character database through the color storage space to obtain the basic character information, and the latest character feature information is updated to the character database, thereby continuously improving the best character feature information in the character database.

[0051] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, various changes, modifications, substitutions and variations of these embodiments are made without departing from the principles and spirit of the present invention, and still fall within the scope of protection of the present invention.

[0052] In the description of the specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0053] The above contents are merely examples and explanations of the present invention. Various modifications or additions to the specific embodiments described or replacements in similar ways by technicians in the technical field shall fall within the protection scope of the present invention as long as they do not deviate from the invention or exceed the scope defined by the claims.

[0054] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for associating basic information of a person based on view data, characterized in that: The basic information association method is as follows: S1. Obtaining view data of the same position through an optical sensor, an optical imaging camera, and a photographic camera, and classifying the view data; S2. Analyze the content in the view according to the classification of the view data and establish three color spaces; In S2, the color space is divided according to the classification status to obtain a view processing space, a color contrast space and a color storage space respectively. The view processing space is used to process view data at the same time, and the color contrast space is used for the system to match and associate the character information on the view with the character information in the color storage space after processing the character information on the view; S3, extracting the character features of infrared imaging in optical imaging, and transmitting the extracted features to the color space, establishing a character framework in the color space according to the character features, and generating a corresponding character feature framework, including the positions of key points in the character features, and analyzing the character posture information to obtain subsequent corresponding character posture data; S4, obtaining the data of S1 and S3, and analyzing the person information in the view captured by the camera, matching the position of the person feature in S3 with the position in the view, and cropping the position of the person in the view; S5, obtaining cropped photo data of the view, and then using an optical sensor to obtain optical data in the current environment, and restoring the color of the person information in the cropped photo data by analyzing the optical data; S6, performing color segmentation on the character information in the view, and extracting the color segmentation and the character information features to the positions of the character feature frames respectively; S7, obtaining the character posture data of S3 and processing the character features in different views, then retrieving the character feature framework in the color space for matching, and using the evaluation network to evaluate the character information in the color space, and finally storing the character feature information with the highest evaluation value in the color space; S8. Obtain character database information, and compare the character database information with the character information in the view.

2. The method for associating basic information of a person based on view data according to claim 1, characterized in that: In the S1, the views captured by the optical imaging and the photographic camera are classified, and the view data at different times are marked and classified. The views of the optical imaging at different times are marked as Aa, Ab...Azz, and the views captured by the photographic camera at different times are marked as Ba, Bb...Bzz. The view data at the same time are retrieved for classification, wherein the character features in the optical imaging views are marked as Aa1, Aa2...Aan, respectively, and the character feature information in the photographic camera views are marked as Ba1, Ba2...Ban, respectively.

3. The method for associating basic information of a person based on view data according to claim 1, characterized in that: In S3, character feature information in the view is obtained, and character data is transmitted to the color space according to the number and characteristics of characters in the view, and a basic framework of the character feature information is established in the view processing space.

4. The method for associating basic information of a person based on view data according to claim 1, characterized in that: In S4, the position in the view captured by the camera is cropped according to the position of the person feature in the view, ensuring that the cropped positions in the same view classification are the same, and further ensuring that the cropped positions in the two views are the same.

5. The method for associating basic information of a person based on view data according to claim 1, characterized in that: In S5, optical data in the optical sensor is retrieved, wherein the optical data includes light intensity and spectrum, and the basic color of the character is restored according to the cropped character information and the optical data, and the color of the overexposed and dark character information is restored.

6. The method for associating basic information of a person based on view data according to claim 1, characterized in that: In S6, after the character's color is restored, the color of the character in the view is segmented and extracted, the character's color with a threshold value within 5 color difference values ​​is extracted, and the extracted color is added to the character feature framework in the color space according to the position of the character's key point.

7. The method for associating basic information of a person based on view data according to claim 1, characterized in that: After the character feature processing of the view is completed in S7, the character feature information stored in the color space is matched with the processed information, and then the evaluation network evaluates the two character feature information to obtain the best character feature information. When the evaluation value is greater than 90%, the color space stops updating the current character feature data.

8. The method for associating basic information of a person based on view data according to claim 1, characterized in that: In the step S8, after acquiring the character database information, the character database information is matched with the basic characteristic information of the character in the color space to acquire the basic character information, and at the same time, the latest characteristic information of the character is updated into the character database.

Images