Method for generating infrared sea ice area by simulating visible light sea ice area

Abstract

The invention relates to a method for generating an infrared sea ice area by simulating a visible light sea ice area, and relates to the technical field of image processing. The method comprises the following steps: determining a skyline in a visible light sea ice image; and dividing the image into a sky image and a sea ice image according to the determined sky line, performing gray scale gradient processing on the sky image, performing simulation processing on the sea ice image, and generating a long-wave infrared imaging effect. According to the method for generating the infrared sea ice area by simulating the visible light sea ice area, an effective infrared sea ice area image can be obtained, and subsequent image fusion and target detection and recognition are facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method for generating an infrared sea ice area by simulating a visible light sea ice area. Background Art

[0002] Infrared radiation is an electromagnetic wave between visible light and microwaves, with a wavelength between 1 and 24 microns, which cannot be directly perceived by the human eye. Objects in nature with a temperature above absolute zero will emit infrared radiation to the outside world. Infrared imaging technology is a technology that uses infrared radiation to detect and process information. This technology converts the infrared radiation emitted by the scene itself into a visible thermal image. Since different parts of an object have different radiation characteristics, the infrared imaging system can distinguish the various parts of the scene and then convert them into visible grayscale images, allowing people to perceive the infrared radiation spectrum that was originally invisible to the naked eye.

[0003] In some scenarios, due to limitations of objective factors such as environmental conditions or onboard equipment, it is very difficult to obtain infrared images in real life. It is particularly difficult to build a large-scale infrared data set, and it is even more difficult to obtain infrared sea ice images in polar sea ice areas. Summary of the invention

[0004] The present invention aims to solve the technical problems in the prior art and provides a method for simulating a visible light sea ice area to generate an infrared sea ice area.

[0005] In order to solve the above technical problems, the technical solutions of the present invention are as follows:

[0006] A method for generating an infrared sea ice area by simulating a visible light sea ice area comprises the following steps:

[0007] Step 1: Determine the skyline;

[0008] Step 2: Divide the image into two parts, sky and sea ice, according to the determined skyline. Perform grayscale gradient processing on the sky part; perform simulation processing on the sea ice part to generate a long-wave infrared imaging effect.

[0009] In the above technical solution, in step 1, Canny edge detection or auxiliary skyline algorithm is adopted to segment the sky and ground objects and determine the skyline.

[0010] In the above technical solution, in step 2, the specific operations of simulating the image of the sea ice portion include: visible light processing, grayscale processing, downsampling processing, upsampling processing, noise addition processing, inversion processing and contrast reduction processing.

[0011] In the above technical solution, the Canny edge detection algorithm in step 1 has the following steps:

[0012] (1) Denoising: Apply Gaussian filtering to smooth the image;

[0013] (2) Gradient: Find the gradient of the image;

[0014] (3) Non-maximum suppression: non-maximum suppression technology is applied to filter out non-edge pixels;

[0015] (4) Apply the double threshold method to determine the possible boundaries;

[0016] (5) Track boundaries using hysteresis techniques.

[0017] In the above technical solution, in step (5), if a weak boundary connected to a certain pixel position and a strong boundary is considered to be a boundary, other weak boundaries are deleted.

[0018] In the above technical solution, the auxiliary skyline algorithm in step 1 has the following steps:

[0019] The skyline was manually drawn with a red line on the sea ice image;

[0020] Convert RGB to HSV color space and define the range of red;

[0021] Find the middle position of each column of red pixels and construct the segmentation line.

[0022] The present invention has the following beneficial effects:

[0023] By using the method proposed in the present invention to simulate the visible light sea ice area and generate the infrared sea ice area, the effective conversion between the visible light image and the infrared image can be realized, and the effective infrared sea ice area image can be obtained under the condition that the optical image of the sea ice area is scarce and the infrared image is missing, so as to enrich the data types and increase the data quantity, lay a solid foundation for the multi-source data fusion and target detection and identification in the ice area environment, and further provide important information guarantee for the safe navigation of ships in the ice area environment. In addition, this method can reduce the cost of obtaining high-quality infrared images and reduce the dependence on expensive equipment; at the same time, it can reduce the dependence on the actual acquisition of data under the harsh environmental conditions of the ice area, and has important application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0025] Figure 1 This is the first type of sea ice background image.

[0026] Figure 2 This is the second sea ice background image.

[0027] Figure 3 The skyline image is detected by the Canny edge detection algorithm.

[0028] Figure 4 The skyline map used by the auxiliary skyline algorithm.

[0029] Figure 5 for Figure 1 The grayscale image is obtained after the image in is grayscaled.

[0030] Figure 6 for Figure 5 The image obtained after downsampling the image in .

[0031] Figure 7 for Figure 6 The image obtained after upsampling the image in .

[0032] Figure 8 for Figure 7 The image is segmented by applying a mask and the part under the sea ice is processed with Gaussian noise.

[0033] Fig. 9 for Figure 8 The image obtained after the image in is inverted.

[0034] Fig.10 In order to perform gradient processing on the upper half of the sea ice image, the image obtained by merging the upper and lower parts of the sea ice image.

[0035] Fig.11 To finally generate an image of part of the simulated infrared sea ice area.

[0036] Fig.12 Flowchart of the image processing process for the lower half of the sea ice. DETAILED DESCRIPTION

[0037] The inventive concept of the present invention is:

[0038] The method of the present invention for simulating visible light sea ice areas to generate infrared sea ice areas is mainly divided into two parts. The first part is to determine the skyline in the visible light sea ice image, and the second part is to apply different methods to the upper and lower parts of the skyline image for image processing respectively.

[0039] The present invention is described in detail below with reference to the accompanying drawings.

[0040] The method of generating an infrared sea ice area by simulating a visible light sea ice area of ​​the present invention comprises the following steps:

[0041] Step 1: Determine the skyline in the visible light sea ice image;

[0042] Use Canny edge detection or auxiliary skyline algorithm to achieve sky-ground segmentation and determine the skyline. When the skyline of some background images is not obvious, the Canny algorithm cannot be used to detect the skyline well. In this case, the auxiliary skyline algorithm needs to be used.

[0043] Figure 1 and Figure 2 These are two different sea ice images. Figure 1 To detect the sea ice image of the skyline using the Canny edge detection algorithm, Figure 2 Sea ice image for detecting the skyline using the skyline-assisted algorithm.

[0044] Specifically, the steps of the Canny edge detection algorithm are as follows:

[0045] (1) Denoising: Apply Gaussian filtering to smooth the image in order to remove noise;

[0046] (2) Gradient: Find the gradient of the image. Calculate the gradient magnitude and direction according to the Sobel operator to find the gradient of the image. First, apply the convolution mask to the x and y directions respectively, and then calculate the gradient magnitude and direction. The formula is as follows:

[0047]

[0048] (3) Non-maximum suppression: Non-maximum suppression technology is used to filter out non-edge pixels and approximate their gradient direction to one of the following values, including 0°, 45°, 90°, 135°, 180°, 225°, 270°, and 315°. The gradient strength of the pixel is compared with the pixels in the positive and negative directions of its gradient. If the pixel has the largest gradient strength, it is retained. Otherwise, it is suppressed, making the blurred boundary clear.

[0049] (4) Apply the double threshold method to determine the possible (potential) boundary, that is, set an upper threshold and a lower threshold. If the pixel in the image is greater than the upper threshold, it is considered to be a boundary; if it is less than the lower threshold, it is considered not to be a boundary; if it is between the two, it is considered to be a candidate and needs to be further processed.

[0050] (5) Hysteresis technology is used to track boundaries. If a certain pixel position and a weak boundary connected to a strong boundary are considered to be boundaries, other weak boundaries are deleted.

[0051] In OpenCV, the Canny() function looks like this:

[0052] edges=Canny(image,threshold1,threshold2[,edges[,apertureSize[,L2gradient]]])

[0053] The parameters involved include:

[0054] image represents the input image;

[0055] Edges represents the output edge map, which has the same size and type as the input image;

[0056] threshold1 represents the first hysteresis threshold;

[0057] threshold2 represents the second hysteresis threshold;

[0058] apertureSize represents the aperture size of the Sobel operator, and its default value is 3;

[0059] L2gradient represents a flag for calculating the gradient magnitude of an image. The default value is false.

[0060] Here, we only need to set two thresholds to detect the skyline well. Figure 1 Convert it to grayscale image gray, edges = cv2.Canny (gray, threshold1 = 50, threshold2 = 15), then traverse each column of the image, find the first edge point, traverse all edge points, connect them into a line, that is, the desired skyline, such as Figure 3 shown.

[0061] like Figure 2 As shown, the skyline in the figure is the skyline obtained according to the auxiliary skyline algorithm.

[0062] The specific steps of the auxiliary skyline algorithm are as follows:

[0063] Manually draw the skyline in red on the sea ice image;

[0064] Convert RGB (red, green, and blue) to HSV (hue (H), saturation (S), brightness (V)) color space to define the range of red;

[0065] Find the middle position of each column of red pixels and construct the segmentation line.

[0066] According to the auxiliary skyline algorithm, Figure 4 As shown, first draw the skyline in red. Figure 4 Then put Figure 4 Convert RGB to HSV color space, define the range of red, in the middle of HSV color, the red range H is 0-10 and 160-179, S is 100-255, V is 100-255, find the middle position of each column of red pixels, and construct the segmentation line, that is, the skyline.

[0067] Step 2: Divide the image into two parts, sky and sea ice, according to the skyline determined in step 1 (upper and lower images respectively), perform grayscale gradient processing on the sky part image, and perform simulation processing on the sea ice part image to generate a long-wave infrared imaging effect.

[0068] Below Figure 1 The background is used as an example to illustrate the specific processing flow. Fig.12 As shown, the following steps are included:

[0069] Grayscale the entire image and apply the function cv2.cvtColor(image,cv2.COLOR_BGR2GRAY). The processing result is as follows Figure 5 As shown;

[0070] Downsample the entire image and apply the function cv2.resize(image,(width / / 2,height / / 2),interpolation=cv2.INTER_LINEAR) to get downsampled_image. The processing result is as follows Figure 6 As shown;

[0071] The entire image is upsampled and the upsampled image is obtained by applying the function cv2.resize(downsampled_image, (width, height), interpolation = cv2.INTER_LINEAR). The processing result is as follows Figure 7 As shown;

[0072] Apply mask segmentation to the entire upsampled image and divide it into two parts, upper_processed and lower_processed. Add Gaussian noise to the sea ice part image and apply the function add_gaussian_noise(lower_processed,var=10) to get noise_image. The processing result is as follows Figure 8 As shown; the add_gaussian_noise function is:

[0073] def add_gaussian_noise(image,mean=0,var=0.01):

[0074] sigma = var

[0075] gauss=np.random.normal(mean,sigma,image.shape)

[0076] noisy_image=np.clip(image+gauss,0,255)

[0077] return noisy_image.astype(np.uint8)

[0078] The sea ice part image is inverted, inverted_image = 255-noise_image, and the processing result is as follows Fig. 9 As shown;

[0079] Perform gradient processing on the sky part of the image. The code is as follows:

[0080] for i in range(width):

[0081] upper_processed[i:]=70*(1+5*i / width)

[0082] Merge the sky and sea ice images, and apply a mask to merge the upper and lower parts together. The processing result is as follows Fig.10 As shown;

[0083] The contrast of the sea ice image is reduced by applying the function

[0084] cv2.convertScaleAbs(image,alpha=1.5,beta=0), the processing result is as follows Fig.11 Shown is the simulated portion of the infrared sea ice area.

[0085] By using the method of the present invention for simulating and generating infrared sea ice areas from visible light sea ice areas, effective infrared sea ice area images can be obtained, which facilitates subsequent image fusion and target detection and recognition.

[0086] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A method for simulating visible light sea ice areas to generate infrared sea ice areas, characterized in that: The following steps are involved: Step 1: Determine the skyline in the visible light sea ice image; Step 2: Divide the image into two parts, sky and sea ice, according to the skyline determined in step 1, perform grayscale gradient processing on the sky part; perform simulation processing on the sea ice part to generate a long-wave infrared imaging effect.

2. The method for generating infrared sea ice area by simulating visible light sea ice area according to claim 1, characterized in that: In step 1, Canny edge detection or auxiliary skyline algorithm is used to segment the sky and ground objects and determine the skyline.

3. The method for generating infrared sea ice area by simulating visible light sea ice area according to claim 1, characterized in that: In step 2, the specific operations of simulating the sea ice part image include: visible light processing, grayscale processing, downsampling processing, upsampling processing, noise adding processing, inversion processing and contrast reduction processing.

4. The method for generating infrared sea ice areas by simulating visible light sea ice areas according to claim 2, characterized in that: The Canny edge detection algorithm in step 1 is as follows: (1) Denoising: Apply Gaussian filtering to smooth the image; (2) Gradient: Find the gradient of the image; (3) Non-maximum suppression: non-maximum suppression technology is applied to filter out non-edge pixels; (4) Apply the double threshold method to determine the possible boundaries; (5) Track boundaries using hysteresis techniques.

5. The method for generating infrared sea ice area by simulating visible light sea ice area according to claim 4, characterized in that: In step (5), if a certain pixel position and a weak boundary connected to a strong boundary are considered to be boundaries, the other weak boundaries are deleted.

6. The method for generating infrared sea ice areas by simulating visible light sea ice areas according to claim 2, characterized in that: The auxiliary skyline algorithm in step 1 has the following steps: The skyline was manually drawn with a red line on the sea ice image; Convert RGB to HSV color space and define the range of red; Find the middle position of each column of red pixels and construct the segmentation line.

Images