WO2020062843A1

WO2020062843A1 - Pill coating image processing method and apparatus, and computer device and storage medium

Info

Publication number: WO2020062843A1
Application number: PCT/CN2019/083136
Authority: WO
Inventors: 陈红丽; 张艳东; 薛绍辰
Original assignee: 河北华讯方舟太赫兹技术有限公司; 深圳市太赫兹科技创新研究院
Priority date: 2018-09-28
Filing date: 2019-04-18
Publication date: 2020-04-02
Also published as: CN109377457A; CN109377457B

Abstract

A pill coating image processing method and apparatus, and a computer device and a storage medium. By means of receiving an original spectrum uploaded by a pill image detection device (302), using multiple types of filtering windows to carry out filtering processing on the original spectrum (304), carrying out Fourier transform on the filtered spectrum to obtain a pill coating image (306), classifying all pixel points of the pill coating image to obtain an impulse noise point set and a stable pixel point set (308), and carrying out de-noising processing on the impulse noise point set and the stable pixel point set to obtain a de-noised pill coating image (310), multiple types of noise introduced by the original spectrum can be eliminated. By means of carrying out de-noising processing on an impulse noise point set and a stable pixel point set, noise of a pill coating image can be eliminated, thus improving the quality of the pill coating image.

Description

Image processing method and device for pill coating, computer equipment and storage medium

Technical field

The present application relates to the field of image processing technology, and in particular, to a method, a device, a computer device, and a storage medium for image processing of a pill coating.

Background technique

With the development of image processing technology, spectral optical coherence tomography has emerged. This imaging technology uses a camera to collect the original spectrum. After the original spectrum is Fourier transformed, a depth image of the object to be measured can be obtained. Fast speed, high signal-to-noise ratio, high resolution, non-invasive measurement, low cost and other advantages. However, in the process of obtaining a pill-coated image by using spectral optical coherence tomography, the quality of the pill-coated image has problems due to the influence of various noises introduced by the original spectrum. Traditional pill-coated image processing methods, in order to eliminate the effects of various noises introduced by the original spectrum, can not completely eliminate the introduction of the original spectrum by subtracting the reference light spectrum or the average value of the original spectrum from the original spectrum. noise. Therefore, eliminating the effects of various noises introduced by the original spectrum and improving the image quality of the pill coating have become technical issues that need to be solved at present.

Summary of the Invention

Based on this, it is necessary to provide a pill coating image processing method, device, computer equipment, and storage medium capable of eliminating various noises introduced by the original spectrum and thereby improving the quality of the pill coating image.

An image processing method for pill coating, the method includes:

Receive the raw spectrum uploaded by the pill image detection equipment;

Performing filtering processing on the original spectrum using multiple filtering windows;

Fourier transform the filtered spectrum to obtain a pill-coated image;

Classifying all pixels of the pill-coated image to obtain a set of impulsive noise points and a set of stationary pixels;

Denoising the impulse noise point set and stationary pixel point set to obtain a pill-coated image after denoising.

In one embodiment, the performing filtering processing on the original spectrum using multiple filtering windows includes:

Performing average filtering on the original spectrum according to the X-direction filtering window to obtain a first spectrum;

Performing average filtering on the original spectrum according to the Y-direction filtering window to obtain a second spectrum;

Average filtering the original spectrum using a square filter window to obtain a third spectrum;

Use the original spectrum, the first spectrum, the second spectrum, the third spectrum, and a preset relationship to generate a filtered spectrum.

In one embodiment, generating the filtered spectrum by using the original spectrum, the first spectrum, the second spectrum, the third spectrum, and a preset relationship includes:

Subtracting the first spectrum and the second spectrum from the original spectrum to obtain an intermediate spectrum;

Use the intermediate spectrum and the third spectrum to obtain a filtered spectrum.

In one embodiment, the impulsive noise point set includes impulsive noise, the stationary pixel point set includes stationary noise, and the denoising processing on the impulsive noise point set and the stationary pixel point set includes:

Removing the impulse noise by a median filtering process;

Performing a curvelet transform on the pill-coated image after removing the impulse noise;

The stationary noise is removed by the curvelet transform.

In one embodiment, after performing the denoising processing on the impulsive noise point set and the stationary pixel point set to obtain a denoised pill-coated image, the method further includes:

Performing edge detection on the de-noised pill-coated image;

Binarize the image of the pill coating after edge detection to obtain the image of the pill coating after removing the background.

A pill-coated image processing device, the device includes:

A spectrum receiving module for receiving the original spectrum uploaded by the pill image detection device;

A filtering processing module, configured to perform filtering processing on the original spectrum by using various filtering windows;

An image generating module for performing a Fourier transform on the filtered original spectrum to obtain a pill-coated image;

A pixel classification module, configured to classify all pixels of the pill-coated image to obtain a set of impulsive noise points and a set of stationary pixels;

A denoising processing module is configured to perform denoising processing on the impulse noise point set and the stationary pixel point set to obtain a decoated pill-coated image.

In one embodiment, the filtering processing module is further configured to perform average filtering on the original spectrum according to the X-direction filtering window to obtain a first spectrum; perform average filtering on the original spectrum according to the Y-direction filtering window to obtain A second spectrum; performing a mean filtering on the original spectrum using a square filter window to obtain a third spectrum; using the original spectrum, the first spectrum, the second spectrum, the third spectrum, and a preset relationship, Generate a filtered spectrum.

In one of the embodiments, the denoising processing module is further configured to remove the impulse noise concentrated in the impulse noise points through a median filtering process; the pill-coated image after removing the impulse noise concentrated in the impulse noise points Performing a curvelet transform; and removing the stationary noise concentrated in the stationary pixels by the curvelet transform.

A computer device includes a memory and a processor. The memory stores a computer program that can run on the processor, and the processor implements the steps in the foregoing method embodiments when the computer program is executed.

A computer-readable storage medium stores a computer program thereon. When the computer program is executed by a processor, the steps in the foregoing method embodiments are implemented.

The above-mentioned pill-coating image processing method, device, computer equipment, and storage medium utilize various filtering windows to filter the original spectrum, thereby eliminating various noises introduced by the original spectrum. The filtered spectrum is used to perform Fourier transform to obtain the pill-coated image. By classifying all the pixels of the pill-coated image, the impulse noise point set and the stationary pixel point set are obtained. Denoising the impulse noise point set and the stationary pixel point set, thereby eliminating the noise of the pill-coated image, thereby improving the quality of the pill-coated image.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is an application environment diagram of a pill coating image processing method in an embodiment; FIG.

2 is a schematic diagram of an original spectrum of an embodiment;

FIG. 3 is a schematic flowchart of a pill coating image processing method in an embodiment; FIG.

4 is a schematic diagram of a pill coating image obtained by performing a Fourier transform on a first spectrum in an embodiment;

5 is a schematic diagram of a pill coating image obtained by performing a Fourier transform on a second spectrum in an embodiment;

6 is a schematic diagram of a pill coating image obtained by performing a Fourier transform on the filtered spectrum in an embodiment;

7 is a schematic diagram of a pill coating image obtained after smooth noise is removed by curve wave transformation in an embodiment;

8 is a structural block diagram of a pill-coated image processing device in an embodiment;

FIG. 9 is an internal structural diagram of a computer device in one embodiment.

detailed description

In order to make the purpose, technical solution, and advantages of the present application clearer, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the application, and are not used to limit the application.

The pill-coating image processing method provided in the present application can be applied to an application environment as shown in FIG. 1. The terminal 104 receives the original spectrum sent by the pill detection device 102, and the original spectrum is shown in FIG. The terminal 104 performs filtering processing on the original spectrum by using various filtering windows. The terminal 104 performs a Fourier transform on the filtered spectrum to obtain a pill-coated image. The terminal 104 classifies all the pixels of the pill-coated image to obtain a set of impulsive noise points and a set of stationary pixels. The terminal 104 performs denoising processing on the impulse noise point set and the stationary pixel point set to obtain a pill-coated image after denoising. The terminal 104 may be various computers, notebook computers, and tablet computers.

In one embodiment, as shown in FIG. 3, a method for processing a pill-coated image is provided. The method is applied to the terminal in FIG. 1 as an example for description, and includes the following steps:

Step 302: Receive the original spectrum uploaded by the pill detection device.

Step 304: Perform filtering processing on the original spectrum by using various filtering windows.

After receiving the original spectrum uploaded by the pill detection device, the terminal can use various filtering windows to filter the original spectrum. The multiple filtering windows include filtering windows of different sizes and filtering windows of different shapes. The filtering windows of different sizes include the X direction. Filter window and Y-direction filter window. Filter windows of different shapes include square filter windows. The filtering process includes mean filtering.

In the traditional image processing method of pill coating, by subtracting the reference light spectrum from the original spectrum, or subtracting the average of all the original spectra from the original spectrum, the various noises introduced by the original spectrum cannot be completely eliminated. In this embodiment, the terminal uses multiple filtering windows to filter the original spectrum, which can eliminate the influence of noise introduced by the original spectrum, improve the ability to eliminate the original spectral noise, and thus help improve the quality of the pill-coated image. .

Step 306: Fourier transform the filtered spectrum to obtain a pill-coated image.

After the terminal performs filtering processing on the original spectrum by using various filtering windows, a Fourier transform can be performed on the filtered spectrum to obtain a pill-coated image. Specifically, the terminal uses the X-direction filter window, the Y-direction filter window, and the square filter window of the various filter windows to filter the original spectrum to obtain a filtered spectrum, and performs Fourier transform on the spectrum to obtain a pill coating. image.

Step 308: Classify all the pixels of the pill-coated image to obtain a set of impulsive noise points and a set of stationary pixels.

After the terminal performs a Fourier transform on the filtered spectrum to obtain a pill-coated image, the terminal can classify all pixels of the pill-coated image to obtain a set of impulsive noise points and a set of stationary pixels. Specifically, the terminal divides all the pixels of the pill-coated image into a set of impulsive noise points and a set of stationary pixels according to the relationship between the central pixel value in the pill-coated image and the pixel values in the neighborhood. The terminal makes a difference between the pixel value of the central pixel point and the pixel values of the surrounding neighborhood to obtain multiple difference values. The terminal classifies all pixels of the pill-coated image according to multiple differences, and obtains a set of impulsive noise points and a set of stationary pixels. The terminal divides the pixel points corresponding to the difference greater than the positive preset value and the value smaller than the negative preset value into a set of impulsive noise points, and the pixel points corresponding to the difference smaller than the positive direction preset value and the difference greater than the negative preset value correspond to The pixels are divided into a set of stationary pixels.

Further, the terminal makes a difference between the pixel value of the central pixel point and the pixel values of the surrounding neighborhood, and when the difference value is greater than or equal to zero, it is the positive direction number of the central pixel point; when the difference value is less than zero, the central pixel point is The number of points in the negative direction. When the number of positive directions is greater than the preset value of the positive direction, it means that the value of the central pixel point is greater than that of most neighboring pixels. The terminal divides the pixels corresponding to the number of positive directions greater than the preset value of the positive direction into a set of impulsive noise points. . When the number of negative directions is less than the preset value of the negative direction, it means that the value of the central pixel point is smaller than most neighboring pixel values. The terminal divides the pixels corresponding to the number of positive directions that are less than the preset value of the negative direction into a set of impulsive noise points. . In both cases, the terminal divides the central pixel point into a set of impulsive noise points. The terminal divides the pixels corresponding to the number of positive directions smaller than the preset value in the positive direction, the pixels corresponding to the number of negative directions greater than the preset value in the negative direction, and the central pixel point into a stationary pixel point set.

Step 310: Denoise the impulse noise point set and the stationary pixel point set to obtain a pill-coated image after denoising.

After the terminal classifies all the pixels of the pill-coated image, and obtains the impulse noise point set and the stationary pixel point set, it can perform denoising processing on the impulse noise point set and the stationary pixel point set to obtain the denoised pill-coated image. Specifically, the terminal performs denoising processing on the set of impulse noise points, and the terminal removes impulse noise from the set of impulse noise points. The terminal performs denoising processing on the stationary pixel set. The stationary pixel set includes normal pixels and stationary noise. The terminal removes the stationary noise concentrated in the stationary pixels to obtain a de-noised pill-coated image.

In this embodiment, a variety of filtering windows are used to perform filtering processing on the original spectrum, so that various noises introduced by the original spectrum can be eliminated. The filtered spectrum is used to perform Fourier transform to obtain the pill-coated image. By classifying all the pixels of the pill-coated image, the impulse noise point set and the stationary pixel point set are obtained. The terminal performs denoising processing on the impulse noise point set and the stationary pixel point set, thereby eliminating the noise of the pill-coated image, thereby improving the quality of the pill-coated image.

In one embodiment, the original spectrum is filtered by using multiple filtering windows, including: performing an average filtering on the original spectrum according to the X-direction filtering window to obtain a first spectrum; and performing an average filtering on the original spectrum according to the Y-direction filtering window to obtain The second spectrum; average filtering the original spectrum using a square filter window to obtain a third spectrum; and generating the filtered spectrum using the original spectrum, the first spectrum, the second spectrum, the third spectrum, and a preset relationship. The terminal filters the original spectrum according to the X-direction filter window, which can remove the high-frequency coherent signals in the original spectrum, obtain the low-frequency noise component, and obtain the first spectrum. As shown in Figure 4, no white lines are present in the pill-coated image. White lines refer to the horizontal lines that exist in the pill-coated image at multiple optical path differences. The X-direction filtering window refers to the X-direction filtering window of the width of the pill-coated image, wherein the X-direction filtering window may be a filtering window of a first preset size. For example, the first preset size may be 1 * 5 (one row and five columns) or 1 * 10 (one row and ten columns).

In the traditional method, the filtering window is used to perform average filtering for the X-direction filtering window of the width of the pill-coated image, and the low-frequency noise component obtained after filtering is inaccurate. Mean filtering, the low-frequency noise component obtained after filtering is more accurate, that is, closer to the real noise.

After the terminal performs average filtering on the original spectrum according to the X-direction filtering window to obtain a first spectrum, the terminal may perform average filtering on the original spectrum according to the Y-direction filtering window to obtain a second spectrum. The pill coating image is shown in FIG. 5, and there are no bright lines in the pill coating image. The bright line refers to the vertical line formed in the image of the pill coating, because the noise floor corresponding to the original spectrum is significantly higher than other spectral lines due to the reflected light of the sample being too strong. The Y-direction filter window may be a filter window of a second preset size. For example, the second preset size may be 10 * 1 (ten rows and one column).

In the traditional way, by subtracting the reference light spectrum from the original spectrum, or subtracting the reference light spectrum from the original spectrum, the bright lines in the pill-coated image cannot be eliminated. In this embodiment, the terminal performs average filtering on the original spectrum according to the Y-direction filter window, which can eliminate bright lines in the pill-coated image.

The terminal performs mean filtering on the original spectrum according to the Y-direction filter window to obtain a second spectrum, and then uses a square filter window to perform mean filtering on the original spectrum to obtain a third spectrum. The terminal obtains a pill obtained by Fourier transform of the filtered spectrum. The coating image is shown in FIG. 6, and there are no white lines and bright lines in the pill coating image. The square filter window may be a third pre-sized filter window. For example, the third preset size may be 10 * 10 (ten rows and ten columns).

After the terminal performs average filtering on the original spectrum using a square filter window to obtain a third spectrum, the terminal may use the original spectrum, the first spectrum, the second spectrum, the third spectrum, and a preset relationship to generate a filtered spectrum. The preset relationship can be addition and subtraction.

In one embodiment, generating the filtered spectrum by using the original spectrum, the first spectrum, the second spectrum, the third spectrum, and a preset relationship includes: subtracting the first spectrum and the second spectrum from the original spectrum to obtain an intermediate spectrum; The intermediate spectrum is added to the third spectrum to obtain a filtered spectrum.

The terminal subtracts the first spectrum from the original spectrum to eliminate white lines in the pill-coated image. The terminal subtracts the spectrum after subtracting the first spectrum and then subtracts the second spectrum to obtain an intermediate spectrum, which can eliminate bright lines in the pill-coated image. The terminal adds the middle spectrum to the third spectrum to remove the low-frequency noise portion of the spectrum after subtracting the first spectrum, and at the same time removes the white and bright lines in the pill-coated image to maintain the balance of the pill-coated image. The effect of eliminating the noise introduced by the original spectrum is achieved, and the white and bright lines in the pill-coated image can be completely removed, further improving the quality of the pill-coated image.

Further, the terminal may also generate an intermediate spectrum by using the original spectrum and a third spectrum, and subtract the first spectrum and the second spectrum by using the intermediate spectrum to obtain a filtered spectrum. That is, when the terminal filters the original spectrum, the processing order of the first spectrum, the second spectrum, and the third spectrum may not be limited. The order of the filtered spectrum obtained by the terminal according to the original spectrum, the first spectrum, the second spectrum, the third spectrum, and a preset relationship can be freely adjusted.

In this embodiment, the terminal may adjust the average filtering order of the original spectrum by using the X-direction filtering window, the Y-direction filtering window, and the square filtering window. The terminal uses the original spectrum, the first spectrum, and the preset relationship to eliminate white lines in the pill-coated image. The terminal uses the original spectrum, the second spectrum, and a preset relationship to eliminate bright lines in the pill-coated image. The terminal uses the original spectrum, the first spectrum, the second spectrum, the third spectrum, and a preset relationship to achieve simultaneous elimination of white lines and bright lines in the pill-coated image and maintain the balance of the pill-coated image.

In one embodiment, the impulse noise point set includes impulse noise, and the stationary pixel point set includes stationary noise. Denoising the impulse noise point set and the stationary pixel point set includes the following steps: removing the impulse noise by a median filtering process; Curved wave transform is performed on the pill-coated image after the pulse noise is removed; stationary noise is removed by the curve wave transform.

In the traditional method, the median filtering process is to sort the pixel values in the filtering window, and use the median to replace the value of the central pixel. It is to process all the pixels in the pill-coated image without difference, and change the details. . However, in this embodiment, the terminal only processes the impulse noise concentrated in the impulse noise points, and does not change the values of other pixels. The filtered pill coating image has higher fidelity, and can retain the basis of pill coating information. This method effectively removes impulse noise in the pill-coated image.

After the terminal removes the impulse noise through the median filtering process, the terminal can perform curve wave transformation on the pill-coated image after the impulse noise is removed. Curvelet transform uses wavelet transform to decompose the image into a series of subband image signals with different scales, and then uses local ridge wave transform to analyze each subband image signal. Curve wave transform is essentially a multi-scale ridge wave transform. This transform has strong directional selectivity for signals with singularities in the curve. The terminal performs curve wave transformation on the pill-coated image after removing the impulse noise to directly obtain a description of high-dimensional features such as straight lines and planes. To realize the smooth noise removal of the pill-coated image, it is necessary to perform curve wave transformation on the pill-coated image to obtain the curve coefficient, and then process the curve coefficient according to a certain threshold function, and finally use the processed curve coefficient to perform the pill package. Clothing image reconstruction.

After the terminal performs curve wave transformation on the pill-coated image after removing the pulse noise, the terminal can remove the stationary noise by curve wave transformation. The edge information of the pill coating image can be well preserved to avoid the lack of edge information leading to the modification of the complete pill coating image. The terminal image of the pill coating obtained by removing the stationary noise through curve wave transformation is shown in Figure 7. The quality of the coated image is obviously better than that of the pill coated image before denoising.

In this embodiment, the terminal removes impulse noise through a median filtering process, performs curve wave transformation on the pill-coated image after removing the pulse noise, and removes smooth noise by curve wave transformation. The noise in the pill-coated image is completely removed, the edge information is completely retained, and the complete pill-coated image is avoided from being modified, thereby further improving the quality of the pill-coated image.

In one embodiment, after performing denoising processing on the impulse noise point set and stationary pixel point set to obtain the denoised pill-coated image, the method further includes: performing edge detection on the denoised pill-coated image; and performing edge detection. The subsequent pill coating image is binarized to obtain a pill coating image after removing the background.

In this embodiment, the terminal obtains the denoised pill coating image, and the pill coating image well retains the edge information of the pill coating image to avoid the lack of edge information leading to the modification of the complete pill coating image. The quality of the pill-coated image is further improved, and the terminal performs edge detection on the denoised pill-coated image to achieve better detection of the edge position of the pill-coated image. The terminal performs a binarization process on the pill-coated image after edge detection, and can more accurately calculate the thickness of the pill-coated image. After the binarization process, the pill-coated image can be obtained after removing the background.

It should be understood that although the steps in the flowchart of FIG. 3 are sequentially displayed in accordance with the instructions of the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless explicitly stated in this document, the execution of these steps is not strictly limited, and these steps can be performed in other orders. Moreover, at least a part of the steps in FIG. 3 may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but may be executed at different times. The execution of these sub-steps or stages The sequence is not necessarily performed sequentially, but may be performed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

In one embodiment, as shown in FIG. 8, a pill-coated image processing device is provided, which includes: a spectrum receiving module 802, a filtering processing module 804, an image generating module 806, a pixel classification module 808, and a denoising processing module. 810 of which:

The spectrum receiving module 802 is configured to receive an original spectrum uploaded by a pill image detection device.

The filtering processing module 804 is configured to perform filtering processing on the original spectrum by using various filtering windows.

An image generation module 806 is configured to perform a Fourier transform on the filtered original spectrum to obtain a pill-coated image.

The pixel classification module 808 is configured to classify all pixels of the pill-coated image to obtain a set of impulsive noise points and a set of stationary pixels.

The denoising processing module 810 is configured to perform denoising processing on the impulse noise point set and the stationary pixel point set, to obtain a denoised pill-coated image.

In one embodiment, the filtering processing module 804 is further configured to perform average filtering on the original spectrum according to the X-direction filtering window to obtain a first spectrum; perform average filtering on the original spectrum according to the Y-direction filtering window to obtain a second spectrum; use a square The filtering window performs average filtering on the original spectrum to obtain a third spectrum; and uses the original spectrum, the first spectrum, the second spectrum, the third spectrum, and a preset relationship to generate a filtered spectrum.

In one embodiment, the filtering processing module 804 is further configured to subtract the first spectrum and the second spectrum from the original spectrum to obtain an intermediate spectrum; and use the intermediate spectrum to add a third spectrum to obtain a filtered spectrum.

In one embodiment, the denoising processing module 810 is further configured to remove the impulse noise by a median filtering process; perform a curvelet transform on the pill-coated image after removing the impulse noise; and remove stationary noise by the curvelet transform.

In one embodiment, a binarization processing module is further included, which is used to perform edge detection on the image of the pill-coated image after denoising; and binarize the image of the pill-coated image after the edge detection to obtain the pill package after removing the background. Clothing image.

Regarding the specific limitation of the pill-coating image processing device, reference may be made to the limitation on the pill-coating image processing method described above, and details are not described herein again. Each module in the above-mentioned pill-coated image processing device may be implemented in whole or in part by software, hardware, and a combination thereof. The above-mentioned modules may be embedded in the hardware form or independent of the processor in the computer device, or may be stored in the memory of the computer device in the form of software, so that the processor calls and performs the operations corresponding to the above modules.

In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. 9. The computer equipment includes a processor, a memory, a network interface, a display screen, and an input device connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for running an operating system and computer programs in a non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. The computer program is executed by a processor to implement a pill-coated image processing method. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a button, a trackball or a touchpad provided on the computer device casing , Or an external keyboard, trackpad, or mouse.

Those skilled in the art can understand that the structure shown in FIG. 9 is only a block diagram of a part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer equipment to which the scheme of the present application is applied. The specific computer equipment may be Include more or fewer parts than shown in the figure, or combine certain parts, or have a different arrangement of parts.

In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the foregoing method embodiments are implemented.

A person of ordinary skill in the art can understand that all or part of the processes in the methods of the foregoing embodiments can be implemented by using a computer program to instruct related hardware. The computer program can be stored in a non-volatile computer-readable storage In the medium, the computer program, when executed, may include the processes of the embodiments of the methods described above. Wherein, any reference to the memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous chain Synchlink DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

The technical features of the above embodiments can be arbitrarily combined. In order to make the description concise, all possible combinations of the technical features in the above embodiments have not been described. However, as long as there is no contradiction in the combination of these technical features, they should be It is considered to be the range described in this specification.

The above-mentioned embodiments only express several implementation manners of the present application, and their descriptions are more specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be noted that, for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can be made, which all belong to the protection scope of the present application. Therefore, the protection scope of this application patent shall be subject to the appended claims.

Claims

An image processing method for pill coating, the method includes:

Receive the raw spectrum uploaded by the pill image detection equipment;

Performing filtering processing on the original spectrum using multiple filtering windows;

Fourier transform the filtered spectrum to obtain a pill-coated image;

Classifying all pixels of the pill-coated image to obtain a set of impulsive noise points and a set of stationary pixels;

Denoising the impulse noise point set and stationary pixel point set to obtain a pill-coated image after denoising.
The method according to claim 1, wherein the filtering the original spectrum by using multiple filtering windows comprises:

Performing average filtering on the original spectrum according to the X-direction filtering window to obtain a first spectrum;

Performing average filtering on the original spectrum according to the Y-direction filtering window to obtain a second spectrum;

Average filtering the original spectrum using a square filter window to obtain a third spectrum;

Use the original spectrum, the first spectrum, the second spectrum, the third spectrum, and a preset relationship to generate a filtered spectrum.
The method according to claim 2, wherein generating the filtered spectrum using the original spectrum, the first spectrum, the second spectrum, the third spectrum, and a preset relationship comprises:

Subtracting the first spectrum and the second spectrum from the original spectrum to obtain an intermediate spectrum;

Use the intermediate spectrum and the third spectrum to obtain a filtered spectrum.
The method according to claim 1, wherein the impulse noise point set includes impulse noise, the stationary pixel point set includes stationary noise, and the denoising is performed on the impulse noise point set and the stationary pixel point set. Processing includes:

Removing the impulse noise by a median filtering process;

Performing a curvelet transform on the pill-coated image after removing the impulse noise;

The stationary noise is removed by the curvelet transform.
The method according to claim 1, wherein after performing the denoising processing on the impulsive noise point set and the stationary pixel point set to obtain a denoised pill-coated image, the method further comprises:

Performing edge detection on the de-noised pill-coated image;

Binarize the image of the pill coating after edge detection to obtain the image of the pill coating after removing the background.
A pill-coated image processing device, wherein the device includes:

A spectrum receiving module for receiving the original spectrum uploaded by the pill image detection device;

A filtering processing module, configured to perform filtering processing on the original spectrum by using various filtering windows;

An image generating module for performing a Fourier transform on the filtered original spectrum to obtain a pill-coated image;

A pixel classification module, configured to classify all pixels of the pill-coated image to obtain a set of impulsive noise points and a set of stationary pixels;

A denoising processing module is configured to perform denoising processing on the impulse noise point set and the stationary pixel point set to obtain a decoated pill-coated image.
The apparatus according to claim 6, wherein the filtering processing module is further configured to perform average filtering on the original spectrum according to the X-direction filtering window to obtain a first spectrum; and perform filtering on the original spectrum according to the Y-direction filtering window. The spectrum is average-filtered to obtain a second spectrum; a square filter window is used to average-filter the original spectrum to obtain a third spectrum; the original spectrum, the first spectrum, the second spectrum, and the third spectrum are used; Spectra and preset relationships to generate filtered spectra.
The device according to claim 6, wherein the denoising processing module is further configured to remove the impulse noise concentrated in the impulse noise point by a median filtering process; The clothing image is subjected to a curvelet transform; the curvelet transform is used to remove stationary noise concentrated in the stationary pixels.
A computer device includes a memory and a processor, and the memory stores a computer program executable on the processor, wherein the processor implements any one of claims 1 to 5 when executing the computer program. Steps of the method.
A computer-readable storage medium having stored thereon a computer program, characterized in that the computer program, when executed by a processor, implements the steps of the method according to any one of claims 1 to 5.