Endoscopic image processing method

Abstract

An endoscopic image processing method is applicable to imaging a affected area using an endoscope. The method includes: providing a first image and a second image of the affected area, wherein the first and second images have different wavelengths or polarization states; performing binarization image processing on the second image; performing positive overlay processing on the second image based on the result of the binarization image processing to form a first positive overlay image; performing pseudo-color processing on the first positive overlay image to form a first pseudo-color processed image; and integrating the first pseudo-color processed image with the first image to form a first integrated image. This endoscopic image processing method acquires images of multiple wavelengths or polarization states at the same location, and improves the endoscopic image quality of the affected area by processing and integrating these images.

Description

Endoscopic image processing methods This invention relates to an image processing method, and more particularly to an endoscopic image processing method. Current clinical endoscopy primarily utilizes visible light imaging, with different diameter endoscopes used depending on the location and intended treatment method. Improving the quality of endoscopic images of affected areas to meet the needs of clinical medical imaging is a research direction in this field. This invention relates to an endoscopic image processing method that acquires images of multiple wavelengths or polarization states at the same location, and improves the endoscopic image quality of the affected area by processing and integrating these images. According to an embodiment of the present invention, an endoscopic image processing method is applicable to imaging a affected area using an endoscope, comprising: providing a first image and a second image of the affected area, wherein the first image and the second image have different wavelengths or polarization states; performing binarization image processing on the second image; performing positive overlay processing on the second image based on the result of the binarization image processing to form a first positive overlay image; performing pseudo-color processing on the first positive overlay image to form a first pseudo-color processed image; and integrating the first pseudo-color processed image with the first image to form a first integrated image. In an embodiment of the present invention, before performing binarization image processing on the second image, the brightness and / or contrast of the second image is further adjusted. In an embodiment of the present invention, before performing pseudo-color processing on the first positive overlay image, the contrast of the first positive overlay image is further adjusted. In an embodiment of the invention, before the first pseudo-color processed image is integrated with the first image, the brightness and / or contrast of the first pseudo-color processed image is further adjusted. In an embodiment of the present invention, the first image is a visible light image, and the second image is a near-infrared light image, a short-wave infrared light image, or a polarized light image. In an embodiment of the present invention, the endoscopic image processing method further includes providing a third image of the affected area, wherein the wavelengths or polarization states of the first image, the second image, and the third image are different; performing binarization image processing on the third image; performing positive overlay processing on the third image based on the result of the binarization image processing on the third image to form a second positive overlay image; performing pseudo-color processing on the second positive overlay image to form a second pseudo-color processed image; and integrating the second pseudo-color processed image, the first pseudo-color processed image, and the first image to form a second integrated image. In an embodiment of the present invention, before performing binarization image processing on the third image, the brightness and / or contrast of the third image is further adjusted. In an embodiment of the present invention, before performing pseudo-color processing on the second positive overlay image, the contrast of the second positive overlay image is further adjusted. In an embodiment of the present invention, before the second pseudo-color processed image, the first pseudo-color processed image, and the first image are integrated, the brightness and / or contrast of the second pseudo-color processed image is further adjusted. In an embodiment of the present invention, the first image is a visible light image, and the second and third images are two of a near-infrared light image, a short-wave infrared light image, and a polarized light image. The endoscopic image processing method of the present invention provides a first image and a second image of a affected area with different wavelengths or polarization states. The second image is binarized. Based on the result of the binarization, the second image is subjected to positive overlay processing to form a first positive overlay image. The first positive overlay image is then subjected to pseudo-color processing to form a first pseudo-color processed image. Finally, the first pseudo-color processed image is integrated with the first image to form a first integrated image. Because this endoscopic image processing method acquires images of multiple wavelengths or polarization states at the same location, the quality of endoscopic images of the affected area can be effectively improved by processing and integrating these images. Reference will now be made in detail to exemplary embodiments of the invention, examples of which are illustrated in the accompanying drawings. Wherever possible, the same element references are used in the drawings and description to denote the same or similar parts. Figure 1 is a schematic flowchart of an endoscopic image processing method according to an embodiment of the present invention. Figure 2 is a block diagram of Figure 1. Figure 3 is an actual image of Figure 2. Please refer to Figures 1 to 3. The endoscopic image processing method 100 of this embodiment is applicable to using an endoscope to photograph the affected area. By integrating images captured by multiple light sources, the differences between the affected area and other tissues are enhanced to assist doctors in making better judgments. The endoscopic image processing method 100 of this embodiment includes the following steps. First, in step 110, a first image 10 and a second image 20 of the affected area are provided, wherein the first image 10 and the second image 20 have different wavelengths or polarization states. For example, the first image 10 is a visible light (VIS) image, and the second image 20 is a near infrared (NIR) image, a short-wave infrared (SWIR) image, or a polarized light image. In this embodiment, the second image 20 is a near-infrared image, ranging from 700 to 1100 nanometers. As shown in Figure 3, the second image 20 is a grayscale image, only presenting signal intensity. Near-infrared light can be used in conjunction with fluorescent developing agents used in the human body. These agents are primarily composed of ICG (indocyanine green). Near-infrared light excites ICG to produce fluorescence. The wavelength of near-infrared light is approximately 780-800 nanometers, while the wavelength of fluorescence is approximately 820 nanometers. Near-infrared light can be used to acquire images through most silicon-based image sensors, and different wavelengths of images can be obtained by using filters of different wavelengths. Due to the properties of optics, the depth to which light can penetrate varies in different tissues of the human body. This is because the different layers of tissue in each organ affect the depth to which photons can penetrate. For example, the mucous membrane of the digestive tract is quite different from the skin on the body surface. For instance, in terms of the penetration depth of light in the skin and mucous membranes (400-1800 nanometers), near-infrared light can obtain deeper tissue images than visible light and short-wave infrared light. Green light, a visible light source, has a wavelength of about 500 nanometers and a penetration depth of about 1 millimeter in mucosal tissue. However, if ICG is excited with near-infrared light at 800 nanometers, its image can be captured at a depth of about 5 millimeters. Since the metabolic rate of ICG in tumors and some tissues is slower than that in normal tissues, a larger amount of ICG will remain in the tumor. This can be used to understand the image of tumors deeper in the mucosa. It can also be used to identify blood vessels by measuring the concentration difference of ICG in blood vessels and tissues. Therefore, the properties of near-infrared light that can be used for ICG fluorescence can help understand the distribution of tumors at deeper levels. In addition, polarized images of visible or near-infrared light can provide more information about the boundaries between the mucosa and the tumor. While short-wave infrared light cannot reach the depth of near-infrared light, nor can it effectively target specific cells or tumors like ICG, it can create a contrasting display by utilizing the absorption characteristics of substances at specific wavelengths without the need for external contrast agents. Furthermore, the application of polarized light imaging has contributed to biological tissues, especially tumors and cardiovascular tissues. Human tissues contain a lot of collagen, and because collagen itself is a non-centrosymmetric material, it will change with polarized light. Therefore, the application of polarized imaging in endoscopy is also valuable. It can help us provide information on the boundaries of tumors and such plaques, so that doctors can better understand the determination of the removal boundary when performing treatments endoscopically. When removing plaques, it can be confirmed whether the blood vessel wall is affected. In other words, designers can choose the appropriate type of light based on the type of affected area to better detect the affected area. The primary image in this case is still the first image 10 (visible light image). The second image 20 (near-infrared light, short-wave infrared light or polarized light image) is superimposed on the first image 10 (visible light image) after being characterized, so as to obtain an integrated image for doctors to more effectively judge the affected area. Of course, in other embodiments, the number of images is not limited to two. For arterial embolism, the first image 10 can be visible light, the second image 20 can be short-wave infrared, and the third image 30 (Figure 5) can be polarized light. For tumor imaging, the first image 10 can be visible light, the second image 20 can be near-infrared, and the third image 30 can be polarized light. For tissue blood leakage, the first image 10 can be visible light, and the second image 20 can be short-wave infrared. The above combinations can yield more favorable images of the affected area for treatment after image processing and integration. Following step 110, the endoscopic image processing method 100 may optionally include step 115, adjusting the brightness and / or contrast of the second image 20. As shown in FIG2, in this embodiment, the brightness and / or contrast of the first image 10 may also optionally be adjusted. Next, in step 120, the second image 20 is binarized. As shown in Figure 3, after binarization, the binarized image 22 is obtained. That is, an image with only black and white, and no grayscale. Next, in step 130, based on the result of the binarized image processing, the second image 20 is subjected to a positive overlay process to form a first positive overlay image 24. In step 130, the white areas in the second image 20 corresponding to the binarized image are retained, and the black areas in the second image 20 corresponding to the binarized image are turned black, so as to form a first positive overlay image 24 with clearer edges. After step 130, the endoscopic image processing method 100 may optionally include step 135, adjusting the contrast of the first positive overlay image 24. Next, in step 140, the first positive overlay image 24 is subjected to pseudo-color processing to form the first pseudo-color processed image 26. As shown in Figure 3, the pseudo-color processing can use colors other than those in the first image 10 (such as blue or yellow-green) to make the affected area more clearly highlighted during subsequent integration. After step 140, the endoscopic image processing method 100 may optionally include step 145, adjusting the brightness and / or contrast of the first pseudo-color processed image 26. Finally, in step 150, the first pseudo-color processed image 26 is integrated with the first image 10 to form the first integrated image 50. As shown in Figure 3, the tumor site in the first integrated image 50 is more clearly marked by light blue. Figure 4 is a flowchart illustrating an endoscopic image processing method according to another embodiment of the present invention. Figure 5 is a block diagram of Figure 4. Referring to Figures 4 and 5, the main difference between the endoscopic image processing method 100a of Figure 4 and the endoscopic image processing method 100 of Figure 1 is that the endoscopic image processing method 100a of Figure 4 integrates three images, and the endoscopic image processing method 100a further includes the following steps. Step 112: Provide a third image 30 of the affected area, wherein the wavelengths or polarization states of the first image 10, the second image 20, and the third image 30 are different. The first image 10 is a visible light image, and the second image 20 and the third image 30 are two of the following: near-infrared light images, short-wave infrared light images, and polarized light images. In the case of arterial embolism, the first image 10 can be visible light, the second image 20 can be short-wave infrared, and the third image 30 can be polarized light. In the case of tumor imaging, the first image 10 can be visible light, the second image 20 can be near-infrared, and the third image 30 can be polarized light. Of course, the types of the first image 10, the second image 20, and the third image 30 are not limited to these. Following step 112, the endoscopic image processing method 100a may optionally include step 117, adjusting the brightness and / or contrast of the third image 30. Next, in step 122, the third image 30 is subjected to binarization image processing. Then, in step 132, based on the result of the binarization image processing of the third image 30, a positive overlay processing is performed on the third image 30 to form a second positive overlay image. Following step 132, the endoscopic image processing method 100a may optionally include step 137, adjusting the contrast of the second positive overlay image. Next, in step 142, pseudo-color processing is performed on the second positive overlay image to form a second pseudo-color processed image. Following step 142, the endoscopic image processing method 100a may optionally include step 147, adjusting the brightness and / or contrast of the second pseudo-color processed image. Finally, in step 160, the second pseudo-color processed image, the first pseudo-color processed image 26, and the first image 10 are integrated to form the second integrated image 60. Similarly, in this embodiment, the main image is the first image 10 (visible light image). The second image 20 and the third image 30 (two of the near-infrared light, short-wave infrared light and polarized light images) are superimposed on the first image 10 (visible light image) after feature processing to obtain an integrated image, which allows doctors to more effectively judge the affected area. The endoscopic image processing method of the present invention provides a first image and a second image of a affected area with different wavelengths or polarization states. The second image is binarized. Based on the result of the binarization, the second image is subjected to positive overlay processing to form a first positive overlay image. The first positive overlay image is then subjected to pseudo-color processing to form a first pseudo-color processed image. Finally, the first pseudo-color processed image is integrated with the first image to form a first integrated image. Because this endoscopic image processing method acquires images of multiple wavelengths or polarization states at the same location, the quality of endoscopic images of the affected area can be effectively improved by processing and integrating these images. Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention. 10: First image; 20: Second image; 22: Binarized image processing image; 24: First positive overlay image; 26: First pseudo-color processed image; 30: Third image; 50: First integrated image; 60: Second integrated image; 100, 100a: Endoscopic image processing method; 110~160: Steps Figure 1 is a flowchart illustrating an endoscopic image processing method according to an embodiment of the present invention. Figure 2 is a block diagram of Figure 1. Figure 3 is an actual image of Figure 2. Figure 4 is a flowchart illustrating an endoscopic image processing method according to another embodiment of the present invention. Figure 5 is a block diagram of Figure 4. 100: Endoscopic Image Processing Methods 110~150: Steps

Claims

1. An endoscopic image processing method, applicable to imaging a affected area using an endoscope, comprising: The method provides a first image and a second image of the affected area, wherein the first image and the second image have different wavelengths or polarization states; performs binarization image processing on the second image; performs positive overlay processing on the second image based on the result of the binarization image processing to make the edges of the second image clearer, thereby forming a first positive overlay image; performs pseudo-color processing on the first positive overlay image to form a first pseudo-color processed image; integrates the first pseudo-color processed image with the first image to form a first integrated image; and wherein, before performing the binarization image processing on the second image, the method further includes: adjusting the brightness and / or contrast of the second image.

2. The endoscopic image processing method as described in claim 1, wherein before performing pseudo-color processing on the first positive overlay image, it further includes: The contrast of the first positive overlay image is adjusted.

3. The endoscopic image processing method as described in claim 1, wherein before the first pseudo-color processed image is integrated with the first image, it further includes: The brightness and / or contrast of the first pseudo-color processed image are adjusted.

4. The endoscopic image processing method as described in claim 1, wherein the first image is a visible light image and the second image is a near-infrared light image, a short-wave infrared light image, or a polarized light image.

5. The endoscopic image processing method as described in claim 1 further includes: A third image of the affected area is provided, wherein the wavelengths or polarization states of the first image, the second image, and the third image are different; the third image is subjected to binarization image processing; based on the result of the binarization image processing of the third image, the third image is subjected to positive overlay processing to form a second positive overlay image; the second positive overlay image is subjected to pseudo-color processing to form a second pseudo-color processed image; and the second pseudo-color processed image, the first pseudo-color processed image, and the first image are integrated to form a second integrated image.

6. The endoscopic image processing method as described in claim 5, wherein before performing the binarization image processing on the third image, it further includes: The brightness and / or contrast of the third image are adjusted.

7. The endoscopic image processing method as described in claim 5, wherein before performing the pseudo-color processing on the second positive film overlay image, it further includes: The contrast of the second positive overlay image is adjusted.

8. The endoscopic image processing method as described in claim 5, wherein before integrating the second pseudo-color processed image, the first pseudo-color processed image, and the first image, the method further includes: The brightness and / or contrast of the second pseudo-color processed image are adjusted.

9. The endoscopic image processing method as described in claim 5, wherein the first image is a visible light image, and the second image and the third image are two of a near-infrared light image, a short-wave infrared light image, and a polarized light image.

Images