CN110492934B - Noise suppression method for visible light communication system - Google Patents

Noise suppression method for visible light communication system Download PDF

Info

Publication number
CN110492934B
CN110492934B CN201910627476.0A CN201910627476A CN110492934B CN 110492934 B CN110492934 B CN 110492934B CN 201910627476 A CN201910627476 A CN 201910627476A CN 110492934 B CN110492934 B CN 110492934B
Authority
CN
China
Prior art keywords
pixel point
target pixel
value
pixels
noise
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910627476.0A
Other languages
Chinese (zh)
Other versions
CN110492934A (en
Inventor
刘宏展
卢国钧
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
South China Normal University
Original Assignee
South China Normal University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by South China Normal University filed Critical South China Normal University
Priority to CN201910627476.0A priority Critical patent/CN110492934B/en
Publication of CN110492934A publication Critical patent/CN110492934A/en
Application granted granted Critical
Publication of CN110492934B publication Critical patent/CN110492934B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
    • H04B10/11—Arrangements specific to free-space transmission, i.e. transmission through air or vacuum
    • H04B10/114—Indoor or close-range type systems
    • H04B10/116—Visible light communication
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/60—Noise processing, e.g. detecting, correcting, reducing or removing noise
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/70—SSIS architectures; Circuits associated therewith
    • H04N25/76—Addressed sensors, e.g. MOS or CMOS sensors

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Optical Communication System (AREA)

Abstract

The invention discloses a noise suppression method of a visible light communication system, which relates to the technical field of visible light communication and comprises the steps of obtaining a line of pixels of a received image in the visible light communication system, wherein the line of pixels is positioned outside an RGB-LED light source halation occupied area in the image; constructing a distinguishing Boolean value, and carrying out stripe classification on the line of pixels based on the distinguishing Boolean value, wherein the distinguishing Boolean value is determined according to the maximum value of the R, G, B channel difference value of two different pixel points and a first threshold; determining the number of pixel points in each type of stripe, and determining a noise stripe according to a second threshold; determining the color of each non-noise stripe to complete data demodulation; the method can greatly reduce the noise of the sample sheet; only a certain line of pixels in the image is studied, and all pixels in the sample are not calculated, so that the calculation speed and the simplicity are improved.

Description

Noise suppression method for visible light communication system
Technical Field
The invention relates to the technical field of visible light communication, in particular to a noise suppression method for a visible light communication system.
Background
Visible Light Communication (VLC) has excellent characteristics such as low cost and absolute safety, and large environment support of Light Emitting Diodes (LEDs) all together, which makes the VLC system a candidate for the next-generation wireless communication technology. In a visible light communication system, there are two receiving methods, one is reception by a photodiode, and the other is reception by a Complementary Metal Oxide Semiconductor (CMOS) image sensor, which is hereinafter referred to as a CMOS-VLC system.
The CMOS image sensor is almost in the camera of all smart phones, and the specific imaging principle makes it possible for users to obtain information from the LED light source through the camera of the smart phone. However, in the process of capturing the optical signal by using the mobile phone camera, various noises are generated, and particularly, under the condition of large bandwidth, the signal-to-noise ratio is greatly reduced, so that the communication quality is sharply reduced. In the process of image processing of the imaging sample, how to effectively design an algorithm to remove noise becomes a key point.
In denoising, there is a small group of studies to accomplish data acquisition by using a low-pass filter, fitting a curve and performing decision decoding by setting a threshold value according to the curve. This method can effectively remove high frequency noise when the bandwidth is low, but when the bandwidth becomes large, a simple low-pass filter will filter out part of the effective information. However, there are many solutions Based on machine learning models recently and all of them are used in visible light communication, and there are groups to study the Noise reduction processing using Density-Based Noise application Spatial Clustering (DBSCAN) algorithm in constellation diagram to solve the problem of amplitude jitter occurring in transmission process.
Disclosure of Invention
The invention provides a noise suppression method for a visible light communication system aiming at the problems of the background technology, which effectively reduces the noise problem of a CMOS-VLC system and has lower complexity than the traditional demodulation method.
In order to achieve the above object, the present invention provides a noise suppression method for a visible light communication system, including the following steps:
acquiring a line of pixels of a received image in a visible light communication system, wherein the line of pixels is positioned outside an area occupied by RGB-LED light sources in the image;
constructing a distinguishing Boolean value, and carrying out stripe classification on the line of pixels based on the distinguishing Boolean value, wherein the distinguishing Boolean value is determined according to the R, G, B channel difference maximum value of two different pixel points and a first threshold;
determining the number of pixel points in each type of stripe, and determining noise stripes according to a second threshold;
and determining the color of each non-noise stripe to complete data demodulation.
Preferably, the constructing a distinguishing boolean value, and performing stripe classification on the line of pixels based on the distinguishing boolean value, wherein the distinguishing boolean value is determined according to the maximum value of the R, G, B channel difference between two different pixel points and a first threshold; the method specifically comprises the following steps:
determining unprocessed target pixel points in the row of pixels;
creating a new cluster of the target pixel point;
traversing the pixel points in the cluster to obtain another unprocessed pixel point around the target pixel point;
determining the maximum value of the difference value of the R, G, B channels of the target pixel point and the other pixel point;
comparing the maximum value of the difference value with a first threshold value, and determining the category relationship between the target pixel point and the other pixel point;
and continuously expanding the clusters of the target pixel points until the clusters cannot be expanded.
Preferably, the maximum value of the difference is compared with a first threshold value, and the category relationship between the target pixel point and the other pixel point is determined; the method specifically comprises the following steps:
comparing and judging whether the maximum value of the difference value is smaller than a first threshold value;
if so, the target pixel point and the other pixel point belong to the same stripe category, and the other pixel point is added into a cluster of the target pixel point;
otherwise, the target pixel point and the other pixel point do not belong to the same stripe category.
Preferably, the continuously expanding the cluster of the target pixel point until the cluster of the target pixel point cannot be expanded again specifically includes:
judging whether all unprocessed pixel points around the target pixel point have been traversed or not; if yes, continuously traversing other target pixel points in the cluster;
judging whether all target pixel points in the complete clusters are traversed or not; if yes, further judging whether all the pixel points are classified, and if yes, completing the classification of the pixel points.
Preferably, the obtaining of another unprocessed pixel point around the target pixel point specifically includes:
and acquiring another unprocessed pixel point within the range of 3-5 pixel points around the target pixel point.
Preferably, the first threshold is set according to brightness of the entire image.
Preferably, the first threshold value ranges from 30 to 40.
The invention provides a noise suppression method for a visible light communication system, which can greatly reduce the noise of a specimen; only a certain line of pixels in the image is studied, and all pixels in the sample are not calculated, so that the calculation speed and the simplicity are improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the structures shown in the drawings without creative efforts.
Fig. 1 is a schematic diagram illustrating a difference between imaging modes of a CCD image sensor and a CMOS image sensor, wherein (a) is a schematic diagram illustrating the CCD image sensor, and (b) is a schematic diagram illustrating the CMOS image sensor;
FIG. 2 is a schematic diagram of the operation of a visible light communication system based on a CMOS image sensor;
FIG. 3 is a schematic view of noise due to the zipper effect;
FIG. 4 is a schematic diagram of an imaging process when the scanning frequency and the transmitting frequency of the transmitting end are not the same;
FIG. 5 is a diagram illustrating noise caused by inconsistency between the scanning frequency and the transmitting frequency of the transmitting end;
FIG. 6 is a diagram illustrating a pseudo-parallel operation process of a transmitting end;
FIG. 7 is a diagram illustrating noise due to pseudo-parallel operation of the transmitter;
FIG. 8 is a flow chart illustrating a method for noise suppression in a visible light communication system according to an embodiment of the present invention;
FIG. 9 is a schematic diagram illustrating one row of pixels in a selected proof in accordance with an embodiment of the present invention;
FIG. 10 is a flowchart detailing the step S20 according to an embodiment of the present invention;
FIG. 11 is a graph showing experimental results in one embodiment of the present invention, wherein (a) is an unprocessed image, (b) is a processed analyzed cluster, and (c) is a processed image;
FIG. 12 is a diagram illustrating a numerical analysis of a green channel of an image according to an embodiment of the present invention, wherein (a) is a green channel distribution diagram of an unprocessed image, and (b) is a green channel distribution diagram of a processed image;
the implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that, if directional indications (such as up, down, left, right, front, and back … …) are involved in the embodiment of the present invention, the directional indications are only used to explain the relative positional relationship between the components, the movement situation, and the like in a specific posture (as shown in the drawing), and if the specific posture is changed, the directional indications are changed accordingly.
In addition, if there is a description of "first", "second", etc. in an embodiment of the present invention, the description of "first", "second", etc. is for descriptive purposes only and is not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In addition, technical solutions between various embodiments may be combined with each other, but must be realized by a person skilled in the art, and when the technical solutions are contradictory or cannot be realized, such a combination should not be considered to exist, and is not within the protection scope of the present invention.
In CMOS-VLC systems, the transmitting end is generally divided into two categories: white light LEDs are used as the transmitting end, and light emitting diodes of three primary colors of red, green and blue (RGB-LEDs) are used as the transmitting end.
The invention mainly adopts a modulation scheme that an RGB-LED light source is used as a sending end and OOK is used as three lights. The specific implementation method is that continuous three bits of the binary sequence are taken, and the binary sequence is loaded on three switches of the RGB-LED to realize the modulation of eight colors. Specific implementation methods and examples are shown in tables 1 and 2;
TABLE 1 binary sequence and color correspondence scheme
Figure BDA0002127583090000041
Table 2: modulation examples
Figure BDA0002127583090000042
At present, most smart phones adopt a CMOS image sensor as a camera instead of a conventional CCD image sensor. The difference between the imaging modes of the CCD image sensor, called Global Shutter (Global Shutter), and the CMOS image sensor, is given as shown in fig. 1 (a) and (b), i.e., each column starts exposure at the same time and ends exposure at the same time and then reads out data. The imaging mode of the CMOS image sensor is called Rolling Shutter (Rolling Shutter), and when imaging, each column starts exposure in sequence, and after exposure, data is read out in sequence and recorded in a picture, and the specific process is shown in fig. 2.
Bayer filters, which is a mosaic color filter sequence formed by arranging RGB filters on a photo-sensing element, are currently carried on almost all CMOS image sensors. This filter is 50% green, 25% blue and the remaining 25% red, and is therefore also referred to as an RGGB filter. Each pixel collected in the CMOS image sensor based on the color filter can only be one of red, green and blue light, and the other two missing primary colors are interpolated by using the color information of the pixels around the primary colorsThe value was obtained. Because the interpolation algorithm is actually a color reconstruction process, almost all interpolation algorithms have distortion phenomena, such as a zipper effect, that is, when an image jumps from a low-frequency region to a high-frequency region, pixels are distributed at regular intervals like a zipper. In a CMOS-VLC system, increasing the bandwidth enables data to be transmitted at a very high frequency, and the optical signal also changes at a very high frequency, which causes a large amount of noise at the edge of the color change due to the zipper effect, thereby causing a large interference to decoding. FIG. 3 shows the speed of transmission at 105The zipper effect occurring at bit/s or more causes the frequency of stripe change to become very large due to excessively fast transmission speed, so that the zipper effect occurs almost on each stripe.
The CMOS image sensor integrates the received light signal during the exposure time, and records the integrated result in a picture as a pixel value at the end of the exposure process.
When the scanning frequency of the CMOS image sensor is consistent with the sending frequency of the sending end, the exposure time t isbIs equal to the transmission time tf. And the initial time of exposure is equal to the initial time of sending, therefore the imaging process of the CMOS image sensor is as shown in figure 2, noise and error code phenomena can not occur, and the CMOS image sensor belongs to ideal imaging;
when the scanning frequency of the CMOS image sensor is not consistent with the sending frequency of the sending terminal, or the exposure initial time is not equal to the sending initial time, the imaging process of the CMOS image sensor is as shown in fig. 4, in the second exposure process, the red light signal is collected in the first half of the time, the blue light signal is collected in the second half of the time, and the violet light signal is read out when the pixel is read out, thereby forming an error code and noise. This is an unavoidable phenomenon because it cannot be guaranteed that the scanning frequency and the transmitting-end transmitting frequency coincide and the exposure initial timing and the transmitting initial timing coincide by one hundred percent. Fig. 5 illustrates this noise phenomenon found during the experiment.
When the RGB-LED light source is lit at the transmitting end, it is necessary to light a plurality of light sources in many cases, for example, red light and blue light are simultaneously lit for lighting purple light. However, in the lighting process, it is often impossible to light two light sources simultaneously, and it is necessary to light red light first and then light blue light (or vice versa), which causes a slight delay, and also in the extinguishing process. As shown in fig. 6, a legend of noise due to pseudo-parallelism is given. Although the proportion of the noise is not large, the noise can still be captured by the high-speed scanning imaging of the CMOS image sensor. Fig. 7 illustrates this noise phenomenon found during the experiment.
The DBSCAN algorithm is a relatively common unsupervised machine learning algorithm, clusters in any shape can be automatically searched according to the distance between data, noise points and outliers can be effectively found, and the DBSCAN algorithm is an excellent and efficient data clustering algorithm. Meanwhile, the DBSCAN algorithm requires two parameters as operation preconditions: eps and MinPts. The former determines the maximum radius range of data points in search of similar data points, and the latter determines the minimum number of data points for which a cluster is not considered as noise. Because the effective data occupies pixel points far larger than the noise in the samples obtained in the CMOS-VLC system, the method is very suitable for noise reduction processing by using a DBSCAN algorithm, but the relationship among the pixels also relates to colors (R, G, B three values) besides the distance;
therefore, in the embodiment of the present invention, a new machine learning DBSCAN algorithm is designed for a CMOS-VLC system on the basis of a conventional DBSCAN algorithm, specifically as follows:
in a first preferred embodiment of the present invention, as shown in fig. 8, the method comprises the following steps:
s10, acquiring a line of pixels of a received image in the visible light communication system, wherein the line of pixels is located outside an area occupied by RGB-LED light sources in the image;
in the embodiment of the invention, in the sample sheet obtained by the CMOS-VLC system, except for the area (the area framed by the white circle on the figure) occupied by the RGB-LED light source halation, the carried information of each line of pixels is the same, so in order to reduce the operation time of DBSCAN, only one line of pixels with complete information in the sample sheet is taken as shown in figure 9;
in the embodiment of the invention, all pixel points in a row of taken pixels are marked as 'unprocessed', each pixel point in the row is processed one by one, and the processed pixel points are marked as 'processed';
s20, constructing a distinguishing Boolean value, and classifying the stripes of the line of pixels based on the distinguishing Boolean value, wherein the distinguishing Boolean value is determined according to the maximum value of the difference value of the R, G, B channels of two different pixel points and a first threshold value;
as shown in fig. 10, specifically:
s201, determining unprocessed target pixel points in the row of pixels;
s202, creating a new cluster of the target pixel points;
in the embodiment of the invention, a first pixel point is selected, a new cluster corresponding to the pixel point is created, and the pixel point is marked as processed;
s203, traversing another unprocessed pixel point around the target pixel point;
s204, determining the maximum value of the difference value of the R, G, B channels of the target pixel point and the other pixel point;
s205, comparing the maximum value of the difference value with a first threshold value, and determining the category relationship between the target pixel point and the other pixel point; the method specifically comprises the following steps: comparing and judging whether the maximum value of the difference value is smaller than a first threshold value; if yes, the target pixel point and the other pixel point belong to the same stripe category, and the other pixel point is added into a cluster of the target pixel point and marked with 'processed'; otherwise, the target pixel point and the other pixel point do not belong to the same stripe category and are not processed;
in the embodiment of the present invention, it is determined whether two pixels (hereinafter, represented by pixel 1 and pixel 2) belong to the same stripe, and a difference boolean value F is proposed, which is determined by the color channel values of pixel 1 and pixel 2 and a set first threshold maxdif:
F(r1，r2，g1，g2，b1，b2，maxdif)＝max[|r1-r2|，|g1-g2|，|b1-b2|]-maxdif (1)
wherein r is1: r channel value, R, of pixel 12: of the pixel 2R channel number, g1: g-channel value, G, of pixel 12: g channel value of pixel 2, b1: b channel value of Pixel 1, B2: the B channel value of pixel 2;
in the embodiment of the invention, the first threshold maxdif needs to be set in advance according to the brightness of the whole sample, and is generally set to be 30-40; when the distinguishing Boolean value F is larger than 0, the color difference of the two pixels is too large, and the two pixels are not the same type of pixels. When the distinguishing Boolean value F is less than 0, the two pixels are considered to be the same type of pixels;
s206, judging whether all unprocessed pixel points around the target pixel point are traversed or not, if so, executing the step S207; otherwise, returning to execute the step S203;
s207, traversing all target pixel points in the cluster (traversed and not repeatedly traversed); judging whether all target pixel points in the cluster have been traversed, if so, executing step S208; otherwise, returning to execute the step S203;
s208, judging whether all the pixel points finish marking, if so, finishing pixel point classification, and if not, returning to execute the step S201:
s30, determining the number of pixel points in each type of stripe, and determining noise stripes according to a second threshold;
in the embodiment of the invention, the resolution of the proof is 1440x 1080 and the width is 1080. In this row of pixels, one data (i.e. one stripe in the original image) usually occupies more than 10 pixel points, and the occupied pixel is determined according to the resolution of the sample and the transmission rate of the data, so in the algorithm, the second threshold MinPts is set to 7, which represents that at least 7 pixels in one cluster can represent a stripe class, and if less than 7 pixels in the second threshold are judged to be noise, the noise needs to be removed;
and S40, determining the color of each non-noise stripe, and finishing data demodulation.
In the embodiment of the invention, samples acquired by a CMOS-VLC system are analyzed, a graph (a) in a graph in FIG. 11 is unprocessed samples, but because the samples contain quite a lot of noise problems, if a low-pass filter is simply used for removing noise, a lot of error codes still occur, after the analysis of a DBSCAN algorithm, a machine can learn respective areas of effective stripes by itself and screen out noise signals, and a graph (b) in FIG. 11 reflects a processing result, wherein black and white stripes represent clusters of the stripes, and gray stripes (which are not very intuitive on the graph due to thinner noise signals) represent the noise stripes. In the diagram (c) in fig. 11, the original image is subjected to sharpening processing by using the classified clusters, so that the algorithm effect becomes more intuitive, and meanwhile, the noise stripes are set to be black, so that after the processing by the DBSCAN algorithm, the stripes become clear, the noise signals are all removed, and the carried information is consistent with the information carried by the original image.
In the embodiment of the present invention, the green channel value of the original image is analyzed, and as shown in fig. 12 (a), it can be seen that the image carries many high-frequency noise signals. However, after the DBSCAN algorithm processing, most of the noise signals of the image are filtered out as shown in the diagram (b) in fig. 12.
In the embodiment of the invention, in order to test the performance of the algorithm, an RGB-LED light source is used as a transmitter, a single chip microcomputer is used for randomly generating a binary sequence and modulating the binary sequence to the RGB-LED light source, and a mobile phone camera is used as a receiver for collecting optical signals and storing the optical signals as pictures. Bit Error Rate (BER) is used as a performance metric. The BER is the error rate of the decoded data bits. And decoding the received sample by using an improved DBSCAN algorithm, and comparing a decoding result with an initially generated binary sequence to obtain an error rate.
TABLE 3 error Rate at different Transmission rates
Table 3 Bit error rate at different transmission speed
Figure BDA0002127583090000081
In the embodiment of the present invention, data transmission rates were respectively tested 102bit/s，103bit/s，104bit/s，105bit/s four cases, see table 3. The experiment was performed with a short transmission distance, RGB-LED to CMOS mapThe image sensor surface is about 2cm, and as the transmission distance increases, the number of bit acquisitions decreases substantially, so no measurement is performed. When the transmission rate is lower than 103At bit/s, no errors were observed. When the transmission rate rises to 104When the bit/s is larger than the threshold value, 2.3X 10 can be obtained-3Error rate of. As the transmission rate increases, the bit error rate also increases. Due to the increase of the transmission rate, the area size of the effective stripe is similar to that of the noise stripe, so the DBSCAN cannot distinguish which kind of timing noise. Thus, the algorithm is considered to be at a transmission rate of 104Operation is good below bit/s.
The invention provides a noise suppression method for a visible light communication system, which can greatly reduce the noise of a specimen; the invention discloses a DBSCAN algorithm which is successfully applied to a CMOS-VLC system for the first time. Experimental results show that the algorithm can effectively reduce noise and has the transmission rate of 104Operation is good below bit/s.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention, and all modifications and equivalents of the present invention, which are made by the contents of the present specification and the accompanying drawings, or directly/indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims (6)

1. A noise suppression method for a visible light communication system is characterized by comprising the following steps:
acquiring a line of pixels of a received image in a visible light communication system, wherein the line of pixels is positioned outside an area occupied by RGB-LED light sources in the image;
constructing a distinguishing Boolean value, and carrying out stripe classification on the line of pixels based on the distinguishing Boolean value, wherein the distinguishing Boolean value is determined according to the R, G, B channel difference maximum value of two different pixel points and a first threshold;
determining the number of pixel points in each type of stripe, and determining noise stripes according to a second threshold;
determining the color of each non-noise stripe to complete data demodulation;
the construction of a distinguishing Boolean value, and the stripe classification of the line of pixels is carried out based on the distinguishing Boolean value, wherein the distinguishing Boolean value is determined according to the R, G, B channel difference maximum value of two different pixel points and a first threshold value; the method specifically comprises the following steps:
determining unprocessed target pixel points in the row of pixels;
creating a new cluster of the target pixel point;
traversing the pixel points in the cluster to obtain another unprocessed pixel point around the target pixel point;
determining the maximum difference value of the R, G, B channels of the target pixel point and the other pixel point;
comparing the maximum value of the difference value with a first threshold value, and determining the category relationship between the target pixel point and the other pixel point;
and continuously expanding the clusters of the target pixel points until the clusters cannot be expanded.
2. The method according to claim 1, wherein the comparing the maximum difference value with a first threshold determines a category relationship between the target pixel point and the another pixel point; the method specifically comprises the following steps:
comparing and judging whether the maximum value of the difference value is smaller than a first threshold value;
if so, the target pixel point and the other pixel point belong to the same stripe category, and the other pixel point is added into a cluster of the target pixel point;
otherwise, the target pixel point and the other pixel point do not belong to the same stripe category.
3. The method for suppressing noise in a visible light communication system according to claim 1, wherein the continuously expanding the cluster of the target pixel point until the cluster of the target pixel point cannot be expanded any more specifically comprises:
judging whether all unprocessed pixel points around the target pixel point have been traversed or not; if yes, continuously traversing other target pixel points in the cluster;
judging whether all target pixel points in the clusters are traversed or not; if yes, further judging whether all the pixel points are classified, and if yes, completing the classification of the pixel points.
4. The method according to claim 1, wherein the obtaining of another unprocessed pixel around the target pixel specifically comprises:
and acquiring another unprocessed pixel point within the range of 3-5 pixel points around the target pixel point.
5. The method of claim 1, wherein the first threshold is set according to brightness of the entire image.
6. The method according to claim 5, wherein the first threshold value ranges from 30 to 40.
CN201910627476.0A 2019-07-12 2019-07-12 Noise suppression method for visible light communication system Active CN110492934B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910627476.0A CN110492934B (en) 2019-07-12 2019-07-12 Noise suppression method for visible light communication system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910627476.0A CN110492934B (en) 2019-07-12 2019-07-12 Noise suppression method for visible light communication system

Publications (2)

Publication Number Publication Date
CN110492934A CN110492934A (en) 2019-11-22
CN110492934B true CN110492934B (en) 2022-05-13

Family

ID=68547025

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910627476.0A Active CN110492934B (en) 2019-07-12 2019-07-12 Noise suppression method for visible light communication system

Country Status (1)

Country Link
CN (1) CN110492934B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111726582B (en) * 2020-06-22 2021-12-14 广东技安科技有限公司 Waterproof outdoor monitoring system and method applying numerical analysis

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080158396A1 (en) * 2006-08-07 2008-07-03 Transchip, Inc. Image Signal Processor For CMOS Image Sensors
CN101673412B (en) * 2009-09-29 2011-07-27 浙江工业大学 Light template matching method of structured light three-dimensional vision system
CN101667303B (en) * 2009-09-29 2013-01-16 浙江工业大学 Three-dimensional reconstruction method based on coding structured light
CN104599273B (en) * 2015-01-22 2017-07-28 南京理工大学 Sea and sky background infrared small target detection method based on multi-scale wavelet crossing operation
CN107923737B (en) * 2015-12-13 2019-12-17 富通尼奥有限责任公司 Method and apparatus for superpixel modulation and ambient light rejection
CN109936713B (en) * 2017-12-19 2020-08-21 陕西外号信息技术有限公司 Method and apparatus for decoding information conveyed by a light source

Also Published As

Publication number Publication date
CN110492934A (en) 2019-11-22

Similar Documents

Publication Publication Date Title
US7800661B2 (en) Programmable pattern matching device
CN105005973B (en) A kind of method and device of the quick denoising of image
US5420971A (en) Image edge finder which operates over multiple picture element ranges
JP3862140B2 (en) Method and apparatus for segmenting a pixelated image, recording medium, program, and image capture device
US8120665B2 (en) Image processing method and apparatus, digital camera, and recording medium recording image processing program
US20090067721A1 (en) Image processing apparatus and method and a computer readable medium having computer executable instructions stored thereon for performing the method
CN111784605B (en) Image noise reduction method based on region guidance, computer device and computer readable storage medium
US20060251322A1 (en) Color fringe desaturation for electronic imagers
CN108174057B (en) Method and device for rapidly reducing noise of picture by utilizing video image inter-frame difference
CN104350744A (en) Camera system with multi-spectral filter array and image processing method thereof
KR20080080107A (en) Automatic removal of purple fringing from images
CN106888355B (en) Bit-rate controller and method for limiting output bit rate
CN112734674A (en) Image denoising method
JP5229328B2 (en) Character area extraction device, imaging device having character area extraction function, and character area extraction program
CN111355936B (en) Method and system for acquiring and processing image data for artificial intelligence
US20110085729A1 (en) De-noising method and related apparatus for image sensor
CN111144337A (en) Fire detection method and device and terminal equipment
CN110492934B (en) Noise suppression method for visible light communication system
US9696470B2 (en) Sensing images and light sources via visible light filters
US20140071309A1 (en) Signal shaping for improved mobile video communication
CN116958058A (en) Lens dirt detection method and device and image detection equipment
US20130142441A1 (en) Information processing apparatus and method
KR20060007901A (en) Apparatus and method for automatic extraction of salient object from an image
CN114998619A (en) Target color classification method, device and computer-readable storage medium
US11288537B2 (en) Image forensics using non-standard pixels

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant