A segment splicing processing method of human infrared image

Abstract

The application relates to the technical field of general image data processing or generation, and particularly relates to a human body infrared image segmentation and splicing processing method. 1 1,2 The coordinates in the image coordinate system corresponding to P1 and the coordinates in the image coordinate system corresponding to P2 are obtained 2 1,2 The coordinates in the image coordinate system corresponding to P1 and the coordinates in the image coordinate system corresponding to P2 are obtained 2 1,2 The coordinates in the image coordinate system corresponding to P1 and the coordinates in the image coordinate system corresponding to P2 are obtained The application can splice the infrared images obtained by two infrared image collectors to obtain a complete human body infrared image.

Description

Technical Field

[0001] The present invention relates to the general field of image data processing or generation technology, and in particular to a segmented splicing processing method for human infrared images. Background Art

[0002] The pixel value of each pixel in an infrared image of the human body represents the temperature at the corresponding location on the body. These temperature values ​​can be used as reference information for assessing a person's health. Infrared image collectors can be used to capture infrared images of the human body. However, when the horizontal distance between the human body and the infrared image collector is fixed, due to the limited field of view of a single infrared image collector, to obtain infrared images of the entire human body, two infrared image collectors are required, positioned vertically and horizontally. Each infrared image collector can only capture a partial infrared image of the human body. The infrared images captured by the two infrared image collectors correspond to different image coordinate systems. Therefore, how to stitch these two images together to obtain a complete infrared image of the human body is an urgent problem to be solved. Summary of the Invention

[0003] The present invention aims to provide a segmented splicing processing method for human infrared images, which can splice infrared images respectively acquired by two infrared image collectors to obtain a complete human infrared image.

[0004] According to the present invention, a segmented mosaic processing method for human infrared images is provided, comprising the following steps:

[0005] S100, using a first infrared image collector to obtain a first infrared image P1 of a first segmented area A1 of a target human body, P1=(p 1,1 , p 1,2 ,…,p 1,M ), p 1,m is the mth pixel in P1, where m ranges from 1 to M and M is the number of pixels in P1.

[0006] S200, using a second infrared image collector to obtain a second infrared image P2 of a second segmented area A2 of the target human body, where A1 and A2 have an overlapping area A2. 1,2 ; P2=(p 2,1 , p 2,2 ,…,p 2,N ), p 2,n is the nth pixel in P2, where n ranges from 1 to N, and N is the number of pixels in P2.

[0007] S300, according to P 1 1,2 The coordinates in the image coordinate system corresponding to P1 and P 21,2 The coordinates in the image coordinate system corresponding to P2 obtain the first conversion relationship when the image coordinate system corresponding to P1 is converted to the image coordinate system corresponding to P2; 1 1,2 A 1,2 The set of pixels in P1, P 2 1,2 A 1,2 The set of pixels in P2.

[0008] S400, concatenating P1 and P2 according to the first conversion relationship, including:

[0009] S410, obtain P 1 1,2 The i-th pixel P 1 1,2,i The pixel value T 1 i , i ranges from 1 to Q, Q is P 1 1,2 The number of pixels in .

[0010] S420, obtain P 2 1,2 Zhong and P 1 1,2,i The pixel value T of the corresponding pixel point 2 i , P 2 1,2 Zhong and P 1 1,2,i The corresponding pixel point and P 1 1,2,i Corresponding to A 1,2 in the same position.

[0011] S430: Obtaining a first pixel value adjustment amount and the second pixel value adjustment amount ΔT2=

[0012] S440, obtain P new Medium 1,m The pixel value of the corresponding pixel point T' 1,m =(T 1,m -ΔT1), P new is the infrared image obtained after stitching, T 1,m For p 1,m The pixel value of P new Medium 1,m The coordinates of the corresponding pixel points are based on p 1,m The coordinates in the image coordinate system corresponding to P1 and the first transformation relationship are obtained.

[0013] S450, get Pnew Medium 2,n The pixel value of the corresponding pixel point T' 2,n =(T 2,n -ΔT2), T 2,n For p 2,n The pixel value, P new Medium 2,n The coordinates of the corresponding pixel points are 2,n The coordinates in the image coordinate system corresponding to P2 are equal.

[0014] Compared with the prior art, the present invention has significant beneficial effects. By means of the above technical solution, the segmented mosaic processing method for human infrared images provided by the present invention can achieve considerable technological advancement and practicality, and has wide industrial application value. It has at least the following beneficial effects:

[0015] The present invention uses a first infrared image collector to obtain an infrared image of a first segmented area of ​​a target human body, and uses a second infrared image collector to obtain an infrared image of a first segmented area of ​​a target human body. Since the first infrared image collector and the second infrared image collector both collect area A of the target human body, 1,2 Infrared image of area A 1,2 The coordinates in the image coordinate systems corresponding to P1 and P2 obtain the conversion relationship when the image coordinate system corresponding to P1 is converted to the image coordinate system corresponding to P2. Based on this conversion relationship, the present invention can obtain the coordinates of the pixel points in P1 in the image coordinate system corresponding to P2, and then realize the splicing of P1 and P2.

[0016] On this basis, the present invention also considers that due to the difference in absolute temperature between the first infrared image collector and the second infrared image collector, it is possible that the pixel values ​​corresponding to the same position of the target human body in the infrared images obtained by the two infrared image collectors are different. In order to avoid the problem of inaccurate relative temperature between pixels in the spliced ​​image caused by this situation, the present invention is based on A 1,2 The first pixel value adjustment amount and the second pixel value adjustment amount are obtained based on the corresponding pixel value differences in P1 and P2. Based on the first pixel value adjustment amount and the second pixel value adjustment amount, the pixel values ​​of the pixel points corresponding to A1 and A2 in the stitched image are adjusted respectively, thereby improving the accuracy of the relative temperature between pixels in the stitched image. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 The present invention provides a flowchart of a segmented mosaic processing method for human infrared images. DETAILED DESCRIPTION

[0019] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0020] According to the present invention, a segmented splicing method for human infrared images is provided. Figure 1 As shown, the following steps are included:

[0021] S100, using a first infrared image collector to obtain a first infrared image P1 of a first segmented area A1 of a target human body, P1=(p 1,1 , p 1,2 ,…,p 1,M ), p 1,m is the mth pixel in P1, where m ranges from 1 to M and M is the number of pixels in P1.

[0022] Those skilled in the art know that any infrared image collector in the prior art falls within the protection scope of the present invention.

[0023] S200, using a second infrared image collector to obtain a second infrared image P2 of a second segmented area A2 of the target human body, where A1 and A2 have an overlapping area A2. 1,2 ; P2=(p 2,1 , p 2,2 ,…,p 2,N ), p 2,n is the nth pixel in P2, where n ranges from 1 to N, and N is the number of pixels in P2.

[0024] According to the present invention, the first infrared image collector and the second infrared image collector are used together to collect infrared images of the target human body. The first infrared image P1 and the second infrared image P2 include the entire area of ​​the target human body.

[0025] As an embodiment, the first infrared image collector and the second infrared image collector are at the same horizontal distance from the target human body, the first infrared image collector is arranged above the second infrared image collector, A1 is a segmented area including the head of the target human body, that is, the upper area of ​​the target human body; A2 is a segmented area including the feet of the target human body, that is, the lower area of ​​the target human body; the overlapping area A1,2 The target area is the middle part of the human body.

[0026] Preferably, the resolutions of P2 and P1 are the same, which facilitates the subsequent splicing of P1 and P2.

[0027] S300, according to P 1 1,2 The coordinates in the image coordinate system corresponding to P1 and P 2 1,2 The coordinates in the image coordinate system corresponding to P2 obtain the first conversion relationship when the image coordinate system corresponding to P1 is converted to the image coordinate system corresponding to P2; 1 1,2 A 1,2 The set of pixels in P1, P 2 1,2 A 1,2 The set of pixels in P2.

[0028] According to the present invention, the setting positions (including the horizontal distance from the human body to be captured and the height from the horizontal plane) and the capture angle of view of the first infrared image collector and the second infrared image collector are pre-set and fixed. After the setting positions and capture angle of view of the first infrared image collector and the second infrared image collector are determined, the overlapping area captured by the first infrared image collector and the second infrared image collector is known. In actual application scenarios, when infrared images of a target human body are captured, the horizontal distance of the target human body from the first infrared image collector and the second infrared image collector is fixed. Therefore, the pixel points at which positions in the first infrared image P1 and the second infrared image P2 are the pixels of the overlapping area captured by the two infrared image collectors remain unchanged, and the correspondence between the pixels in the overlapping area of ​​the first infrared image P1 and the pixels in the overlapping area of ​​the second infrared image P2 is also unchanged. Therefore, the above correspondence can be obtained and stored by artificial pre-experimentation.

[0029] According to the present invention, the image coordinate system corresponding to P1 and the image coordinate system corresponding to P2 are both plane coordinate systems. 1 1,2 The coordinates of any pixel point in the image coordinate system corresponding to P1 and P 2 1,2 In the case of the coordinates of the corresponding pixel point in the image coordinate system corresponding to P2, those skilled in the art know that any method in the prior art for obtaining the conversion relationship between the two plane coordinate systems based on the coordinates of the same point in two plane coordinate systems falls within the scope of protection of the present invention. 1 1,2 A pixel point and P 2 1,2The corresponding pixel point in the image is used to obtain the above conversion relationship, and the conversion relationship is used as the first conversion relationship; or according to P 1 1,2 Multiple pixels and P 2 1,2 The above conversion relationship is obtained for the corresponding pixel points in , and the average value of the conversion relationship is used as the first conversion relationship.

[0030] According to the present invention, when the first infrared image collector and the second infrared image collector are arranged in a vertical direction, and the y-axis of P1 and P2 is in a vertical direction, P 1 1,2 and P 2 1,2 The methods for obtaining include:

[0031] S310, traverse P1, if p 1,m The pixel points belong to the target human body, and p 1,m If the y-axis coordinate of p is in the first preset range, 1,m Append to P 1 1,2 Otherwise, do not 1,m Append to P 1 1,2 ;P 1 1,2 is initialized to Null.

[0032] S320, traverse P2, if p 2,n The pixel points belong to the target human body, and p 2,n If the y-axis coordinate of p is in the second preset range, 2,n Append to P 2 1,2 Otherwise, do not 2,n Append to P 2 1,2 ;P 2 1,2 is initialized to Null.

[0033] According to the present invention, the first preset range is related to the conversion relationship between the camera coordinate system of the first infrared image collector and the image coordinate system corresponding to P1, the installation position and acquisition angle of the first and second infrared image collectors, and the horizontal distance of the target human body from the first and second infrared image collectors. Optionally, the first preset range is manually set, and the first preset range can be obtained manually through preliminary experiments.

[0034] According to the present invention, the difference between the pixel points corresponding to the background area and the target human body area in P1 is large, and a threshold value can be set to determine whether the pixel point in P1 is the pixel point corresponding to the target human body area; those skilled in the art know that any threshold value acquisition method in the prior art falls within the scope of protection of the present invention. The present invention determines whether the pixel point in P1 is the pixel point corresponding to the target human body area. 1,m Methods for determining whether a pixel belongs to the target human body include:

[0035] S311, obtain p in P1 1,m The pixel value T 1,m .

[0036] S312, if T 1,m If the pixel value is greater than the preset threshold, it is determined that p 1,m The pixel belongs to the target human body; otherwise, judge p 1,m Pixels that do not belong to the target person.

[0037] According to the present invention, the second preset range is related to the conversion relationship between the camera coordinate system of the second infrared image collector and the image coordinate system corresponding to P2, the installation position and acquisition angle of the first infrared image collector and the second infrared image collector, and the horizontal distance of the target human body from the first infrared image collector and the second infrared image collector. The second preset range is set manually and can be obtained manually by conducting experiments in advance. The present invention determines p 2,n Method and judgment of whether the pixel belongs to the target human body 1,m The method for determining whether a pixel belongs to the target human body is the same and will not be repeated here.

[0038] According to the present invention, the setting of the installation position and the acquisition viewing angle of the first infrared image collector and the second infrared image collector will change the overlapping area A. 1,2 The proportion of the target human body area, preferably, the installation position and acquisition viewing angle of the first infrared image collector and the second infrared image collector should be set to meet the corresponding A when the height of the target human body changes. 1,2 The proportion in the target human body area is 10%-20%, and the resulting splicing effect is better.

[0039] S400: Concatenate P1 and P2 according to the first conversion relationship.

[0040] The present invention uses a first infrared image collector to obtain an infrared image of a first segmented area of ​​a target human body, and uses a second infrared image collector to obtain an infrared image of a first segmented area of ​​a target human body. Since the first infrared image collector and the second infrared image collector both collect area A of the target human body, 1,2 Infrared image of area A 1,2The coordinates in the image coordinate systems corresponding to P1 and P2 obtain the conversion relationship when the image coordinate system corresponding to P1 is converted to the image coordinate system corresponding to P2. Based on this conversion relationship, the present invention can obtain the coordinates of the pixel points in P1 in the image coordinate system corresponding to P2, and then realize the splicing of P1 and P2.

[0041] According to the present invention, splicing P1 and P2 according to the first conversion relationship includes:

[0042] S410, obtain P 1 1,2 The i-th pixel P 1 1,2,i The pixel value T 1 i , i ranges from 1 to Q, Q is P 1 1,2 The number of pixels in .

[0043] S420, obtain P 2 1,2 Zhong and P 1 1,2,i The pixel value T of the corresponding pixel point 2 i , P 2 1,2 Zhong and P 1 1,2,i The corresponding pixel point and P 1 1,2,i Corresponding to A 1,2 in the same position.

[0044] According to the present invention, the correspondence between the pixel points in the overlapping area of ​​the first infrared image P1 and the pixel points in the overlapping area of ​​the second infrared image P2 is constant and can also be manually obtained in advance. The set of pixel points in the human body area in the overlapping area of ​​the first infrared image P1 is P 1 1,2 , the set of human body area pixel points in the overlapping area of ​​the second infrared image P2 is P 2 1,2 , then P 2 1,2 Which pixel is the same as P 1 1,2,i The correspondence is also constant and can be manually pre-obtained.

[0045] S430: Obtaining a first pixel value adjustment amount and the second pixel value adjustment amount

[0046] S440, obtain P new Medium 1,mThe pixel value of the corresponding pixel point T' 1,m =(T 1,m -ΔT1), P new is the infrared image obtained after stitching, T 1,m For p 1,m The pixel value of P new Medium 1,m The coordinates of the corresponding pixel points are based on p 1,m The coordinates in the image coordinate system corresponding to P1 and the first transformation relationship are obtained.

[0047] S450, get P new Medium 2,n The pixel value of the corresponding pixel point T' 2,n =(T 2,n -ΔT2), T 2,n For p 2,n The pixel value, P new Medium 2,n The coordinates of the corresponding pixel points are 2,n The coordinates in the image coordinate system corresponding to P2 are equal.

[0048] The present invention also considers that due to the difference in absolute temperature between the first infrared image collector and the second infrared image collector, it is possible that the pixel values ​​corresponding to the same position of the target human body in the infrared images obtained by the two infrared image collectors are different. In order to avoid the problem of inaccurate relative temperature between pixels in the spliced ​​image caused by this situation, the present invention is based on A 1,2 The first pixel value adjustment amount and the second pixel value adjustment amount are obtained based on the corresponding pixel value differences in P1 and P2. Based on the first pixel value adjustment amount and the second pixel value adjustment amount, the pixel values ​​of the pixel points corresponding to A1 and A2 in the stitched image are adjusted respectively, thereby improving the accuracy of the relative temperature between pixels in the stitched image.

[0049] Although some specific embodiments of the present invention have been described in detail by way of example, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present invention. It should also be understood by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the present invention. The scope of the present invention is defined by the appended claims.

Claims

1. A segmented splicing processing method for human infrared images, characterized in that: The following steps are involved: S100, using a first infrared image collector to obtain a first infrared image P1 of a first segmented area A1 of a target human body, P1=(p 1,1 , p 1,2 ,…,p 1,M ), p 1,m is the mth pixel in P1, where m ranges from 1 to M, and M is the number of pixels in P1; S200, using a second infrared image collector to obtain a second infrared image P2 of a second segmented area A2 of the target human body, where A1 and A2 have an overlapping area A2. 1,2 ; P2=(p 2,1 , p 2,2 ,…,p 2,N ), p 2,n is the nth pixel in P2, where n ranges from 1 to N, and N is the number of pixels in P2; S300, according to P 1 1,2 The coordinates in the image coordinate system corresponding to P1 and P 2 1,2 The coordinates in the image coordinate system corresponding to P2 obtain the first conversion relationship when the image coordinate system corresponding to P1 is converted to the image coordinate system corresponding to P2; 1 1,2 A 1,2 The set of pixels in P1, P 2 1,2 A 1,2 The set of pixels in P2; S400, concatenating P1 and P2 according to the first conversion relationship, including: S410, obtain P 1 1,2 The i-th pixel P 1 1,2,i The pixel value T 1 i , i ranges from 1 to Q, Q is P 1 1,2 The number of pixels in ; S420, obtain P 2 1,2 Zhong and P 1 1,2,i The pixel value T of the corresponding pixel point 2 i , P 2 1,2 Zhong and P 1 1,2,i The corresponding pixel point and P 1 1,2,i Corresponding to A 1,2 the same position in S430: Obtaining a first pixel value adjustment amount and the second pixel value adjustment amount S440, obtain P new Medium 1,m The pixel value of the corresponding pixel point T' 1,m =(T 1,m -ΔT1), P new is the infrared image obtained after stitching, T 1,m For p 1,m The pixel value of P new Medium 1,m The coordinates of the corresponding pixel points are based on p 1,m The coordinates in the image coordinate system corresponding to P1 and the first transformation relationship are obtained; S450, get P new Medium 2,n The pixel value of the corresponding pixel point T' 2,n =(T 2,n -ΔT2), T 2,n For p 2,n The pixel value, P new Medium 2,n The coordinates of the corresponding pixel points are 2,n The coordinates in the image coordinate system corresponding to P2 are equal.

2. The method for segmented splicing of human infrared images according to claim 1, wherein: The first infrared image collector and the second infrared image collector are at the same horizontal distance from the target human body. The first infrared image collector is arranged above the second infrared image collector. A1 is a segmented area including the head of the target human body, and A2 is a segmented area including the feet of the target human body.

3. The method for segmented splicing of human infrared images according to claim 2, wherein: The resolution of P2 and P1 is the same.

4. The method for segmented splicing of human infrared images according to claim 2, wherein: In S300, P 1 1,2 The methods for obtaining include: S310, traverse P1, if p 1,m The pixel points belong to the target human body, and p 1,m If the y-axis coordinate of p is in the first preset range, 1,m Append to P 1 1,2 Otherwise, do not 1,m Append to P 1 1,2 ;P 1 1,2 is initialized to Null.

5. The method for segmented splicing of human infrared images according to claim 4, wherein: In S300, P 2 1,2 The methods for obtaining include: S320, traverse P2, if p 2,n The pixel points belong to the target human body, and p 2,n If the y-axis coordinate of p is in the second preset range, 2,n Append to P 2 1,2 Otherwise, do not 2,n Append to P 2 1,2 ;P 2 1,2 is initialized to Null.

6. The method for segmented splicing of human infrared images according to claim 5, characterized in that: The first preset range and the second preset range are obtained manually through experiments.

7. The method for segmented splicing of human infrared images according to claim 4, wherein: Determine p 1,m Methods for determining whether a pixel belongs to the target human body include: S311, obtain p in P1 1,m The pixel value T 1,m ; S312, if T 1,m If the pixel value is greater than the preset threshold, it is determined that p 1,m The pixel belongs to the target human body; otherwise, judge p 1,m Pixels that do not belong to the target person.

8. The method for segmented splicing of human infrared images according to claim 1, wherein: A 1,2 It covers 10%-20% of the total target human body area.

Images