JPH06337933A - Picture processor - Google Patents

Abstract

PURPOSE:To obtain a picture of high quality emphasizing only the edge of an original picture without emphasizing a noise component included in the original picture by executing weak edge emphasis in a low luminance part and executing strong edge emphasis in a high luminance part. CONSTITUTION:This picture processor is provided with a non-linear circuit 1 for executing the non-linear processing of an original picture and then executing I/O conversion, an edge emphasizing circuit 2 for emphasizing the edge of the non-linearly processed picture and an adder 3 for adding the edge- emphasized picture to the original picture. The circuit 1 compresses the low luminance part of the original picture, extends its high luminance part and converts I/O characteristics into non-linearity. The circuit 2 executes the emphasis of the edge of the non-linearly converted picture. Thereby weak edge emphasis is executed in the low luminance part, strong edge emphasis is executed in the high luminance part and a component expressing the real edge is emphasized. When the edge-emphasized picture is applied to the original picture, only the edge of the original picture is emphasized.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus for carrying out processing for enhancing an edge of an image, particularly an X-ray image.

[0002]

2. Description of the Related Art Conventionally, an X-ray fluoroscopic image has been subjected to image edge enhancement processing in order to be useful for medical diagnosis.

The process of enhancing the edges of an image is performed by enhancing high frequency components in the spatial frequency of the image.

This spatial frequency enhancement processing is typically performed by
For example, the degree of edge enhancement can be changed by performing a template process (Laplacian filter process) that causes a Laplace arithmetic element to act on each pixel of the entire screen, and changing the frequency band to be emphasized or changing the degree of emphasis. .

[0005]

However, conventionally, regardless of the density value of an image, a uniform weight is applied to the entire image to simply emphasize a component having a high spatial frequency. There was a problem of being emphasized. That is, in the X-ray image, there are a low-luminance portion having a large X-ray absorption such as a skeleton and a diaphragm and a high-luminance portion having a small X-ray absorption such as a soft tissue and a lung field.

At the portion where the X-ray absorption is large, S / S is small because there are few X-ray quanta that pass through the subject and there are many scattered X-rays.
Looking at the decrease in the N ratio, conversely, in the portion where the X-ray absorption is small, the S / N ratio is high because there are many X-ray quanta that pass through the subject. When uniform edge enhancement processing is performed on the low-brightness portion having a low S / N ratio and the high-brightness portion having a high S / N ratio, noise in the low-brightness portion is emphasized, and the S / N ratio is further reduced. When viewed as a whole, the image quality is inferior.

In view of the above, the present invention suppresses the enhancement of the noise component contained in the original image and enhances only the edges of the original image, that is, weak edge enhancement is performed in a low luminance portion having a low S / N ratio, and S An object of the present invention is to provide an image processing device capable of performing strong edge enhancement in a high-luminance portion having a high / N ratio.

[0008]

In order to achieve the above object, an image processing apparatus of the present invention includes a non-linear circuit that performs non-linear processing of an original image and performs input / output conversion, and edge enhancement that enhances edges of the non-linearly processed image. A circuit and an addition circuit for adding the edge-enhanced image to the original image are provided, and the non-linear circuit is configured to compress the low-luminance portion of the original image and expand the high-luminance portion.

[0009]

In the original image, the low brightness portion with a low S / N ratio and the S / N
A high-luminance portion having a high N ratio is included, and a low-luminance portion includes a lot of noise components. The non-linear circuit compresses the low-luminance portion (low-luminance signal) of the original image and compresses the high-luminance portion (high-luminance signal) to convert the input / output characteristics to non-linear. The edge enhancement circuit performs edge enhancement on the image subjected to the non-linear conversion processing.

Therefore, weak edge enhancement (weak spatial frequency enhancement) is performed in the low-luminance portion, noise components are suppressed, and strong edge enhancement (strong spatial frequency enhancement) is performed in the high-luminance portion to obtain a true edge. The components represented are highlighted. By adding the edge-enhanced image to the original image in this way, only the edges of the original image are enhanced, and an image in which the noise is suppressed and the image quality is not deteriorated is obtained.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing an embodiment, and FIG. 2 is image data of each part of FIG. In FIG. 1, 1
Is an input / output that non-linearly converts the input / output characteristics so that an original image such as an X-ray fluoroscopic image obtained from a television camera is input, the image signal in the low-luminance portion is compressed, and the image signal in the high-luminance portion is expanded. A non-linear circuit that functions as a converter, and is composed of, for example, a look-up table (LUT).

Reference numeral 2 denotes an edge enhancement circuit for edge-enhancing a non-linearly processed image. For example, a two-dimensional edge is extracted by performing a template process (Laplacian filter process) in which a Laplace arithmetic element is applied to each pixel of the entire screen. The image signal is obtained. An adder circuit 3 adds the edge-enhanced image and the original image.

Next, the operation of FIG. 1 will be described in relation to the image data of FIG. When the image data A in FIG. 2 is input to the non-linear circuit 1, the LUT converts the signal in the low-luminance portion to a smaller size and converts the signal in the high-luminance portion to a more expanded image data B in FIG. The edge enhancement circuit 2 performs edge enhancement processing to obtain a signal of an image from which a two-dimensional edge has been extracted.

At this time, the output image data of the edge enhancement circuit 2 is the image data B of FIG. 2 in which the input data of the same circuit 2 is the image signal of the low luminance portion compressed and the high luminance portion is expanded.
Therefore, the amplitude is small in the low-luminance portion and large in the high-luminance portion, in other words, the weak edge enhancement is performed in the low-luminance portion and the strong edge enhancement is performed in the high-luminance portion.
Image data C.

The image data C of FIG. 2 processed in this way
Is sent to the adding circuit 3 and added to the original image (image data A in FIG. 2). As a result, the output image D of FIG. 2 with the edges emphasized is obtained.

As is clear from the figure, the contrast of the calculation output image D is such that the contrast of the original image is preserved,
The low-brightness portion is weakly edge-emphasized, and the high-luminance portion is strongly edge-enhanced, and only the edges of the original image are emphasized, resulting in an image in which noise is suppressed.

Although the image obtained from the television camera is used as the original image in the above embodiment, it may be stored in the image memory. Also, by detecting the black peak (peak of the low brightness portion) of the original image and automatically changing the non-linear data written in the LUT from this value, optimum edge enhancement is always performed regardless of the inspection site. It can be carried out.

[0018]

According to the image processing apparatus of the present invention, since weak edge enhancement is performed in a low-luminance portion and strong edge enhancement is performed in a high-luminance portion, noise components included in the original image are not emphasized, and edges of the original image are not enhanced. It is possible to obtain a high-quality image with only emphasis.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

FIG. 2 is a diagram showing image data of each part of FIG.

[Explanation of symbols]

1 ... Nonlinear circuit 2 ... Edge enhancement circuit
3 ... Adder circuit

Claims (1)

[Claims] 1. A non-linear circuit that performs non-linear processing so that a low-luminance signal is compressed and a high-luminance signal is expanded with respect to an original image, an edge enhancement circuit that performs edge enhancement processing of the non-linearly processed image, and edge enhancement. An image processing apparatus comprising: an addition circuit that adds the generated image to the original image.