A method for generating pseudo-color eagle-eye code of unpowered weak target infrared measurement image

Abstract

The application discloses a kind of unpowered weak small target infrared measurement image pseudo-color hawk eye coding generation method, including observation area setting, image gray gradient processing, noise elimination, gray range statistics, bilateral pseudo-color coding etc., both highlight the image texture of hot spot area, while also taking into account the visual effect of background area, realize the hot spot attention and pseudo-color enhancement to the area where weak small target is.

Description

Technical Field

[0001] This disclosure relates to the field of image measurement technology, specifically to a method for generating pseudo-color encoding of infrared measurement images of small, unpowered targets. Background Technology

[0002] Currently, most pseudo-color encoding is based on thermal metal encoding and rainbow encoding. Among them, thermal metal encoding is more suitable for infrared radiation images of unpowered, high-speed moving targets in the air. However, this method is completely unsuitable for infrared measurement images of small, unpowered targets with high contrast. This is because the target has no power and its infrared radiation is very weak, resulting in a very small imaging range and a very low grayscale level. Meanwhile, the background area has a wide grayscale range and a large pixel ratio, making it difficult to locate the target area as a hot spot. As a result, the display effect is reduced after the area of ​​interest is converted from a grayscale image to a color image. Summary of the Invention

[0003] In view of this, this disclosure provides a method for generating pseudo-color encoding of infrared measurement images of weak, unpowered targets, which can achieve the best pseudo-color effect in the hotspot area and allows the hotspot area to be set manually, thus realizing the pseudo-color eagle-eye enhancement effect of infrared measurement images of the target range.

[0004] The method for generating pseudo-color eagle-eye encoding of infrared measurement images of small, unpowered targets disclosed herein includes the following steps:

[0005] Step S1: Divide the image encoding area into a target area, an extended target area, and a background area using the eagle eye center and a set radius, where the eagle eye center is movable;

[0006] Step S2: Perform gradient processing on the entire image in both the x and y directions to obtain two gradient images;

[0007] Step S3: Areas with broken grayscale values ​​and a small number of pixels are removed as noise.

[0008] Step S4: Calculate the grayscale range of the three regions respectively, enhance the target region, and compress the grayscale range of the other two regions.

[0009] Step S5: Based on traditional hot metal encoding, the grayscale range of the image is divided into multiple continuous intervals. The middle interval is the strongest and the adjacent intervals on both sides decrease in succession. The image is then projected onto the primary color of the RGB three primary colors. These intervals are then enhanced or weakened according to the visual contrast with the primary color and projected onto the other two colors of the RGB three primary colors respectively.

[0010] This generates a pseudo-color code for the image.

[0011] Furthermore, in step S2, gradient processing is performed on the entire image in two directions as follows:

[0012] x-direction:

[0013]

[0014] y direction:

[0015]

[0016] Where f(x,y) is the grayscale value of the original image, and G x (x,y) and G y (x, y) are the image gradient values ​​in the x and y directions, respectively.

[0017] Furthermore, the noise removal method in step S3 includes:

[0018] If: G x (x,y)-G x (xi,yj)>k、G t (x,y)-G t (xi,yj)>k;

[0019] Then: f(x,y)=f(xi,yj).

[0020] Furthermore, the specific method of step S4 includes:

[0021] Let the grayscale ranges of the target region, the extended target region, and the background region be g, respectively. rb1 g re1 g rb2 g re2 g rb3 g re3 The gray levels of the three regions are enhanced or compressed using coefficients k1, k2, and k3, respectively, where:

[0022] The expansion coefficient k1 of the target region can be set manually, and is generally greater than 1;

[0023] The expansion coefficients k2 and k3 of the other two compression regions are as follows:

[0024]

[0025]

[0026] Where g all This represents the grayscale range of all pixels in the image.

[0027] Furthermore, step S5 specifically includes the following steps:

[0028] Based on traditional hot metal coding, the equalized image grayscale values ​​are divided into 7 intervals, corresponding to the grayscale levels: l0~l1, l1~l2, l2~l3, l3~l4, l4~l5, l5~l6, l6~l7;

[0029] The transformation functions for projecting each interval onto the RGB primary colors are as follows:

[0030]

[0031]

[0032]

[0033] Where R(x,y), G(x,y), and B(x,y) correspond to the three primary colors of red, green, and blue, respectively, and f(x,y) is the gray value of the original image. The gray range is divided into equal parts by default.

[0034] The method disclosed herein draws on eagle-eye technology, sets up a movable target area, enhances or reduces the grayscale of weak target areas and background, and then uses bilateral pseudo-color coding to achieve hot spot focus and pseudo-color enhancement of the area where weak targets are located.

[0035] Compared with the prior art, the beneficial effects of this disclosure are: ① It adopts bilateral pseudo-color coding to achieve pseudo-color enhancement of the hotspot area where weak targets are located; ② The hotspot area can be set and moved; ③ The algorithm is simple and practical. Attached Figure Description

[0036] Figure 1 A flowchart illustrating an exemplary embodiment of this disclosure;

[0037] Figure 2 This is an example of the color projection results for different gray levels. Detailed Implementation

[0038] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0039] This invention provides a method for generating pseudo-color encoding of infrared measurement images of small, unpowered targets, as shown in the appendix. Figure 1 As shown, an exemplary embodiment of this disclosure includes the following steps:

[0040] Step S1, Method for setting the observation area

[0041] First, set the center and radius of the eagle eye. The radius is set as the target radius area and the extended radius area. The center of the eagle eye is the mouse movement position. At this time, the target is the hypothetical target area. The eagle eye can move freely within the entire image area. The image encoding is divided into three parts: the target area r1, the extended target area r2, and the background area r3.

[0042] Step S2, Image grayscale gradient processing

[0043] Gradient processing is applied to the entire image in both the x and y directions to obtain two gradient images, as shown in the following formula:

[0044] G x (x,y)=H(x+1,y)-H(x-1,y)

[0045] G y (x,y)=H(x,y+1)-H(x,y-1)

[0046] In practice, convolution operations are performed on each pixel of the image according to the following template, with a 3×3 convolution kernel:

[0047] Templates related to the x-direction:

[0048]

[0049] Templates related to the y-direction:

[0050]

[0051] Where f(x,y) is the grayscale value of the original image, and G x (x,y) and G y (x, y) are the image gradient values ​​in the x and y directions, respectively.

[0052] Step S3, Noise Removal

[0053] Since infrared sensors are a type of radiation imaging, and generally have a wide radiation range but a narrow effective signal, infrared radiation images of unpowered, high-speed moving targets in the air are unlikely to exhibit the large discontinuities in grayscale values ​​between pixels that are present in visible light sensors. Therefore, areas with grayscale value breaks and a small number of pixels can be removed as noise, as shown in the following formula:

[0054] If: G x (x,y)-G x (xi,yj)>k、G y (x,y)-G y (xi,yj)>k,

[0055] Then: f(x,y)=f(xi,yj), k is the threshold.

[0056] Step S4, Grayscale range statistics

[0057] Calculate the grayscale range g of the three regions respectively. rb1 g re1 g rb2 g re2 g rb3 g re3 The grayscale ranges of the three regions inevitably overlap. To improve the Eagle Eye enhancement effect, the grayscale ranges of the target regions are adjusted accordingly. rb1 g re1 Following the principle of priority, the expansion coefficient k1 of the target region can be manually set, generally greater than 1. The expansion coefficients k2 and k3 of the other two compressed regions are as follows:

[0058]

[0059]

[0060] Step S5, bilateral pseudo-color coding

[0061] Currently, pseudo-color coding is mainly based on hot metal coding and rainbow coding. Among them, hot metal coding is more suitable for infrared radiation images of unpowered, high-speed moving targets in the air. However, its coding is basically based on one-sided coding, that is, the highest gray level corresponds to the edge of the pseudo-color coding. Thus, for weak targets with small gray level gradients, their gray levels cannot be located on the gray level boundaries. Therefore, this disclosure proposes a two-sided coding method, which performs pseudo-color coding on the upper and lower gray levels of the hot spot gray level, as follows:

[0062] Based on traditional hot metal coding, the grayscale region of the grayscale image is divided into seven intervals, corresponding to the following grayscale levels: l0~l1 (blue), l1~l2 (green), l2~l3 (yellow), l3~l4 (red), l4~l5 (yellow), l5~l6 (green), and l6~l7 (blue). The coding transformation function is as follows:

[0063]

[0064]

[0065]

[0066] Where R(x,y), G(x,y), and B(x,y) correspond to the three primary colors of red, green, and blue, respectively, the equalized grayscale level is divided into 7 intervals. Each part is a default interval, but can be freely adjusted as needed. l3 to l4 (red) are used as the hotspot area grayscale encoding, i.e., the eagle eye center, to achieve optimal pseudo-color enhancement for this part. A projection example is attached. Figure 2As shown.

[0067] The pseudo-color eagle-eye coding method for infrared measurement images of weak targets provided in this disclosure proposes a bilateral pseudo-color coding method by setting an eagle-eye region. This method highlights the image texture of the hot spot area while taking into account the visual effect of the background area, thus achieving hot spot focus and pseudo-color enhancement in the area where the weak target is located.

[0068] In summary, the above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for generating pseudo-color eagle-eye encoding for infrared measurement images of small, unpowered targets, characterized in that, Includes the following steps: Step S1: Divide the image encoding area into a target area, an extended target area, and a background area using the eagle eye center and a set radius, where the eagle eye center is movable; Step S2: Perform gradient processing on the entire image in both the x and y directions to obtain two gradient images; Step S3: Areas with broken grayscale values ​​and a small number of pixels are removed as noise. Step S4: Calculate the grayscale range of the three regions respectively, enhance the target region, and compress the grayscale range of the other two regions. Step S5: Based on traditional hot metal encoding, the grayscale range of the image is divided into multiple continuous intervals. The middle interval is the strongest and the adjacent intervals on both sides decrease in succession. The image is then projected onto the primary color of the RGB three primary colors. These intervals are then enhanced or weakened according to the visual contrast with the primary color and projected onto the other two colors of the RGB three primary colors respectively. This generates a pseudo-color code for the image; The specific method of step S4 includes: Let the grayscale ranges of the target region, the extended target region, and the background region be respectively... , , , , , The gray levels of the three regions are respectively calculated using coefficients. k 1 、k 2 、k 3. Enhance or compress, where: Expansion coefficient of the target region Manually set, greater than 1; The expansion coefficients of the other two compression regions As shown in the following formula: in This represents the grayscale range of all pixels in the image. Step S5 specifically includes the following steps: Based on traditional hot metal coding, the equalized image grayscale values ​​are divided into 7 intervals, corresponding to the following grayscale levels: ~ blue, ~ green, ~ yellow, ~ red, ~ yellow, ~ green, ~ blue; The transformation functions for projecting each interval onto the RGB primary colors are as follows: in, , , They correspond to the three primary colors: red, green, and blue. The grayscale values ​​are the original image values. The grayscale range is divided into equal parts by default.

2. The generation method as described in claim 1, characterized in that, In step S2, gradient processing is performed on the entire image in two directions as follows: x-direction: y direction: in, The original image grayscale values, and These are the image gradient values ​​in the x and y directions, respectively.

3. The generation method as described in claim 2, characterized in that, The noise removal method in step S3 includes: if: , but: , k The threshold value is used.

Images