CN204463227U - An Image Filter Processing Accelerator - Google Patents

Abstract

The utility model relates to image processing equipment field, provide a kind of image filtering processor accelerator, comprise gradation of image detection module, data processing module, data transmission module and graphics transport module, described gradation of image detection module is connected with the input port of data processing module, and described data transmission module and graphics transport module are all connected with the delivery outlet of data processing module.By detecting and the gray-scale value of each pixel of record object picture and coordinate figure, calculate normal intensity value ranges, and compare with the actual grey value of pixel, find out the pixel of actual grey value exception and record the coordinate of this pixel, again by these coordinates transmission to filter processing, target image is transferred to filter processing simultaneously, filter processing is according to the coordinate received, directly the pixel of gray-scale value exception is processed, and computing and process need not be carried out to each pixel, save the processing time, improve image filtering treatment effeciency.

Description

An Image Filter Processing Accelerator

technical field

The utility model relates to the field of image processing equipment, in particular to an image filter processing accelerator.

Background technique

Due to the imperfection of imaging system, transmission medium and recording equipment, digital images are often polluted by various noises in the process of formation, transmission and recording. In addition, in some aspects of image processing, when the input object is not as expected, noise will be introduced into the resulting image. These noises often appear as an isolated pixel point or pixel block that causes a strong visual effect on the image. Generally, the noise signal is irrelevant to the object to be studied, and it appears in the form of useless information, disturbing the observable information of the image. For digital image signals, the noise table is a large or small extreme value. These extreme values act on the real gray value of the image pixel through addition and subtraction, causing bright and dark point interference in the image, which greatly reduces the image quality. It will affect the follow-up work such as image restoration, segmentation, feature extraction, and recognition.

In order to eliminate the above noise, that is, to filter the image, the commonly used methods are median filter, mean filter and morphological filter. The pixels are not polluted and do not need to be filtered. Therefore, the existing image filters have done a lot of useless work, which affects the processing speed and efficiency of the image.

Utility model content

The technical problem to be solved by the utility model is to provide an image filter processing accelerator which improves the image filter processing efficiency.

The technical solution adopted by the utility model to solve the technical problem is: an image filter processing accelerator, including an image grayscale detection module, a data processing module, a data transmission module and an image transmission module, the image grayscale detection module and data processing The input port of the module is connected, and the data transmission module and the image transmission module are connected to the output port of the data processing module;

The image grayscale detection module is used to detect the grayscale values of all pixels of the target image, record the coordinates and corresponding grayscale values of each pixel, and transmit the coordinates and grayscale values of each pixel to the data processing module ;

The data processing module is used to compare the gray value of each pixel, determine the normal gray value range, and record the coordinates of the pixel whose gray value exceeds the normal range;

The data transmission module is used to transmit the coordinates of the pixels whose gray value exceeds the normal range to the image subsequent processing device;

The image transmission module is used to transmit the target picture to the image subsequent processing device.

Further, the image grayscale detection module includes a centering unit and a detection unit, the centering unit is used to lock the center point of the target image, and the detection unit uses the center point as the detection starting point and coordinate origin, detects and records each The gray value and coordinates of each pixel.

Further, there are four detection units, which simultaneously detect and record gray values and coordinates of pixels in four quadrants.

Further, a lossless compression module connected to the image transmission module is provided, and the lossless compression module is used to perform lossless compression on the image and transmit it by the image transmission module.

Further, the data processing module is provided with a data storage unit and a data comparison unit, the data storage unit is used to store the gray values and coordinates of all pixels of the previous target image, and the data comparison unit is used to compare the current The gray value of the pixel point with the same coordinates in the target image and the previous target image is calculated, and the gray value difference is calculated, so as to judge whether the pixel point of the current target image at the corresponding coordinate is polluted by noise.

The beneficial effects of the utility model are: by detecting and recording the gray value and coordinate value of each pixel point of the target picture, the data processing module analyzes and calculates the normal gray value range in a certain coordinate area, and compares with this Compare the actual gray value of the pixels within the range, find out the pixel whose actual gray value is too large or too small and record the coordinates of this pixel, and then transmit the coordinates of the pixel with abnormal gray value to the filter processing device At the same time, the target image is transmitted to the filter processing device, and the filter processing device directly processes the pixels with abnormal gray values according to the received coordinates, without having to calculate and process each pixel, which greatly saves the processing time , which improves the image filtering processing efficiency.

Description of drawings

Fig. 1 is a schematic diagram of the principle of the image filter processing accelerator of the present invention.

Detailed ways

Below in conjunction with accompanying drawing and embodiment the utility model is further described.

As shown in Figure 1, a kind of image filter processing accelerator of the present utility model comprises image grayscale detection module 1, data processing module 2, data transmission module 4 and image transmission module 3, described image grayscale detection module 1 and data The input port of the processing module 2 is connected, and the data transmission module 4 and the image transmission module 3 are connected to the output port of the data processing module 2;

The image grayscale detection module 1 is used to detect the grayscale values of all pixels of the target image, record the coordinates and corresponding grayscale values of each pixel, and transmit the coordinates and grayscale values of each pixel to data processing module 2;

The data processing module 2 is used to compare the gray value of each pixel, determine the normal gray value range, and record the coordinates of the pixel whose gray value exceeds the normal range;

The data transmission module 4 is used to transmit the coordinates of the pixels whose gray value exceeds the normal range to the image subsequent processing device;

The image transmission module 3 is used to transmit the target picture to the image subsequent processing device.

Use the image grayscale detection module 1 to detect and record the grayscale value and coordinate value of each pixel of the target picture, and then analyze it by the data processing module 2. The data processing module 2 establishes a three-dimensional coordinate system, and uses the x coordinate and y coordinate as the detection The coordinates of each pixel point, the z coordinate is used as the gray value of the pixel point, and the discrete surface is formed by the gray value of all the pixel points, and the pixel point with a larger or smaller z coordinate value is selected and recorded according to the smoothness of the surface Coordinates, and then transmit the coordinates of the pixels with abnormal gray values to the image subsequent processing device, and at the same time transmit the target image to the image subsequent processing device, the image subsequent processing device is the filter processing device, and the filter processing device directly The pixels with abnormal gray values are processed instead of calculating and processing each pixel, which greatly saves processing time and improves the efficiency of image filtering.

The image grayscale detection module 1 includes a centering unit 11 and a detection unit 12, the centering unit 11 is used to lock the center point of the target image, and the detection unit 12 uses the center point as the detection starting point and the coordinate origin to detect and Record the gray value and coordinates of each pixel. Taking the center of the image as the coordinate origin, the image is divided into four quadrants, which is convenient for orderly detection of pixels.

There are four detection units 12, which simultaneously detect and record the gray values and coordinates of pixels in four quadrants. The gray value and coordinate value detection and recording of the pixel points in the four quadrants are respectively completed by the four detection units 12, the detection time is shortened to a quarter of the original, the detection speed is accelerated, and the detection efficiency is improved.

A lossless compression module 5 connected to the image transmission module 3 is provided, and the lossless compression module 5 is used to perform lossless compression on the image and transmit it by the image transmission module 3 . After the image is losslessly compressed, it is transmitted to the image filter processor to prevent the image from being polluted by secondary noise during the transmission process, thereby affecting the filtering effect.

The data processing module 2 is provided with a data storage unit 21 and a data comparison unit 22, the data storage unit 21 is used to store gray values and coordinates of all pixels of the previous target image, and the data comparison unit 22 is used for Compare the gray value of the pixel with the same coordinates between the current target image and the previous target image and calculate the gray value difference, so as to judge whether the pixel at the corresponding coordinate of the current target image is polluted by noise. For multiple images taken continuously, due to the same or similar shooting environment, equipment and transmission environment, the noise pollution of two adjacent images may be consistent or similar. By comparing the gray value difference of the pixel points of two adjacent images at the same coordinates, it can be judged whether the noise pollution of the two is similar or consistent. If the noise pollution of the two is similar or consistent, the current image needs to be filtered The position of the processed pixel is consistent with the position of the pixel that needs to be filtered in the previous image, and the data processing module 2 directly uses the coordinates of the pixel that needs to be filtered in the previous image as the coordinates of the pixel that needs to be filtered in the current image , and then sent to the image filter processor, which can save a lot of data processing time and improve image processing efficiency.

Claims (5)

1. an image filtering processor accelerator, it is characterized in that: comprise gradation of image detection module, data processing module, data transmission module and image transmission module, described gradation of image detection module is connected with the input port of data processing module, and described data transmission module and image transmission module are all connected with the delivery outlet of data processing module;

Described gradation of image detection module for detecting the gray-scale value of all pixels of target image, record each pixel coordinate and and the gray-scale value of correspondence, and the coordinate of each pixel and gray-scale value are transferred to data processing module;

Described data processing module, for contrasting the gray-scale value of each pixel, determines normal intensity value ranges, and record gray-scale value exceeds the coordinate of the pixel of normal range;

Described data transmission module is used for the coordinates transmission of pixel gray-scale value being exceeded normal range to pictures subsequent treating apparatus;

Described image transmission module is used for Target Photo to be transferred to pictures subsequent treating apparatus.

2. a kind of image filtering processor accelerator as claimed in claim 1, it is characterized in that: described gradation of image detection module comprises centering unit and detecting unit, described centering unit is used for the central point of lock onto target image, described detecting unit, using central point as detection starting point and true origin, detects and records gray-scale value and the coordinate of each pixel.

3. a kind of image filtering processor accelerator as claimed in claim 2, is characterized in that: described detecting unit is four, detects simultaneously and record gray-scale value and the coordinate of the pixel in four quadrants.

4. a kind of image filtering processor accelerator as claimed in claim 1, is characterized in that: be provided with the lossless compression modules be connected with image transmission module, and described lossless compression modules is used for image being carried out Lossless Compression and being transmitted by image transmission module.

5. a kind of image filtering processor accelerator as claimed in claim 1, it is characterized in that: described data processing module is provided with data storage cell and Data Comparison unit, described data storage cell is for storing gray-scale value and the coordinate of all pixels of last target image, described Data Comparison unit for contrasting the gray-scale value of the current target image pixel identical with last target image coordinate and calculating gray-scale value difference, to judge whether the pixel that current target image is positioned at corresponding coordinate is subject to noise pollution.