CN111325696A - Underwater acoustic image reverberation suppression method based on normal distribution interval estimation - Google Patents
Underwater acoustic image reverberation suppression method based on normal distribution interval estimation Download PDFInfo
- Publication number
- CN111325696A CN111325696A CN202010139790.7A CN202010139790A CN111325696A CN 111325696 A CN111325696 A CN 111325696A CN 202010139790 A CN202010139790 A CN 202010139790A CN 111325696 A CN111325696 A CN 111325696A
- Authority
- CN
- China
- Prior art keywords
- image
- target
- normal distribution
- target image
- pixel points
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 230000001629 suppression Effects 0.000 title claims abstract description 22
- 238000003384 imaging method Methods 0.000 claims abstract description 16
- 238000004458 analytical method Methods 0.000 claims description 7
- 238000001914 filtration Methods 0.000 claims description 7
- 208000024891 symptom Diseases 0.000 claims 1
- 230000008569 process Effects 0.000 abstract description 5
- 238000001514 detection method Methods 0.000 abstract description 4
- 238000012800 visualization Methods 0.000 abstract description 4
- 238000012545 processing Methods 0.000 description 5
- 238000006073 displacement reaction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 238000012360 testing method Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G06T5/70—
-
- G06T5/90—
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Abstract
The invention relates to an underwater sound image reverberation suppression method based on normal distribution interval estimation, which is characterized in that all pixel points contained in an original image obtained by high-frequency imaging sonar detection are classified according to a minimum distance to form random variables approximately meeting normal distribution (or unilateral normal distribution) after the pixel points of an extension pixel point of a target image are classified, the generalized center is formed by intrinsic pixel points of the target, the significance level is set, the standard deviation (mean square error) is solved, the normal distribution interval estimation method is applied, interference bright spots caused by reverberation in the original image can be obviously filtered, the problem of more interference in the target discrimination process is solved, and the visualization level of the underwater sound image is improved.
Description
Technical Field
The invention belongs to the field of sonar, and particularly relates to an underwater acoustic image reverberation suppression method based on normal distribution interval estimation.
Background
The phased array three-dimensional acoustic camera sonar (the working frequency is 300kHz) is used as a three-dimensional high-frequency imaging sonar, is a real-time image sonar which displays an underwater target detection result in the form of an underwater sound image, and has the working characteristics that: high working frequency, short action distance, more interference factors in water and large image distortion. Therefore, in the test and application process of the phased array three-dimensional acoustic camera sonar, volume reverberation formed by the anisotropic scattering of fine plankton and tiny bubbles to sound waves and interface reverberation caused by the vertical and horizontal shaking of carriers appear as a large number of discrete and random image bright spots on the formed underwater sound image, thereby causing the following two problems:
(1) image bright spots formed by reverberation additionally increase the image real-time processing workload of UI software, and the processing effect of conventional image noise reduction methods (such as median filtering, weighted smooth filtering and the like) is poor, so that the performance of high-frequency imaging sonar is influenced to a certain extent;
(2) image bright spots formed by reverberation form much interference on target discrimination, and the probability of false discrimination and false alarm is obviously improved.
Therefore, in the process of processing the underwater acoustic image, effective filtering must be carried out on the reverberation image bright spots, and the practical application efficiency of the high-frequency imaging sonar is improved.
Disclosure of Invention
The invention aims to solve the problem of interference of high-frequency imaging sonar in the aspect of target discrimination under the reverberation condition in the prior art, and provides a normal distribution interval estimation-based underwater acoustic image reverberation suppression method.
The invention adopts the following specific technical scheme:
a normal distribution interval estimation-based underwater acoustic image reverberation suppression method comprises the following steps:
s1: carrying out image connected domain analysis algorithm analysis on the original image of the high-frequency imaging sonar to determine the number n of targets in the original image and each target imageCorresponding set of eigenpixel points Ai，i＝1,2,...,n；
S2: based on the corresponding intrinsic pixel point set A of each target imageiSolving the generalized center e of each target imagei，i＝1,2,...,n；
S3: aiming at all the non-target image intrinsic pixel points in the original image, calculating the generalized center e of each pixel point to n target imagesiAnd dividing the target image into the intrinsic pixel point set of the target image corresponding to the nearest generalized center, so that each target image obtains the extended pixel point set A'iFinishing the classification of the intrinsic pixel points of the non-target image;
s4: for each extended pixel point set A'iCalculating each pixel point in the set to the generalized center e of the target imageiAnd constructing a set D with the distance as an elementi；
S5: for each set DiPerforming confidence interval estimation on the elements in the normal distribution, and removing an extended pixel point set A'iAnd (4) filtering the reverberation bright points of the underwater sound image estimated based on the normal distribution interval by taking the removed extended pixel point set as a final intrinsic pixel point set corresponding to the target image according to the sample pixel points of which the distance from the mid-to-generalized center is outside the confidence interval.
Preferably, the generalized central form is a mathematical expression ax + by + cz of the three-dimensional space where the target image is located is 1, and the coefficient of the mathematical expression is solved by using a least square method, that is, the target image a is obtainediGeneralized center e ofiGeneralized center eiA least squares estimate representing a mathematical expression describing sample pixel points of a target image.
Preferably, the non-target image intrinsic pixel points are
Representing the set of all but target images in the original image
And the other pixel points are not.
Preferably, said set DiApproximately satisfying a normal distribution or a unilateral normal distribution.
Preferably, in S5, the set D is referred toiAll elements in (2) are calculated for their standard deviation σiAnd set significance level αiTo a confidence level of 1- αiConfidence interval of, then from A'iEliminating pixel points corresponding to elements outside the confidence interval to obtain a new subset
As the final set of native pixel points for the corresponding target image.
Further, the final eigen pixel points of all target images are collected
All the elements in (2) are drawn on a display interface to serve as a target underwater sound image for reverberation suppression.
Compared with the prior art, the invention has the following beneficial effects:
according to the method, all pixel points contained in an original image obtained by high-frequency imaging sonar detection are classified into random variables approximately meeting normal distribution (or unilateral normal distribution) after finishing the extension pixel points of the target image according to the minimum distance, the intrinsic pixel points of the target form a generalized center, the significance level is set, the standard deviation is solved, the normal distribution interval estimation method is used, interference bright spots caused by reverberation in the original image can be obviously filtered, the problem of more interference in the target discrimination process is solved, and the visualization level of the underwater acoustic image is improved.
Drawings
Fig. 1 is an original display image obtained by detecting the sea bottom with a phased array three-dimensional acoustic imaging sonar.
Fig. 2 is a submarine target image processed by the underwater acoustic image reverberation suppression method according to the present invention when the significance level is set to 0.5.
Fig. 3 is a submarine target image processed by the underwater acoustic image reverberation suppression method according to the present invention when the significance level is set to 0.32.
Detailed Description
The invention will be further elucidated and described with reference to the drawings and the detailed description. The technical features of the embodiments of the present invention can be combined correspondingly without mutual conflict.
In the test and application process of the phased array three-dimensional acoustic camera sonar, the influence of fine plankton, micro-bubbles, the uneven structure of a water body and the like in water on reverberation formed by scattering of incident sound waves is received, and the performance of the sonar is influenced to a certain degree, so that large deviation is caused for target judgment. The influence of reverberation on the performance of the phased array three-dimensional acoustic camera sonar is that the interference in a target image is obviously increased, and the target image is represented in the form that a large number of discrete, random, tiny and unfixed bright spots appear near the target image.
The invention concept of the invention is as follows: firstly, determining the number of targets in an original underwater sound image obtained by high-frequency imaging sonar and the intrinsic pixel points of each target image, solving the generalized center of each target image, finishing the classification of the intrinsic pixel points of the non-target images and obtaining an extended pixel point set of each target image; aiming at each target image expansion pixel point set, establishing a set consisting of the displacement from each sample pixel point to the generalized center of the target image, carrying out confidence interval estimation according to normal distribution, eliminating the sample pixel points of which the displacement from the generalized center in the expansion pixel point set falls outside the confidence interval, and finishing the filtering of the reverberation bright points of the underwater sound image based on the normal distribution interval estimation; the method solves the problem of high-frequency imaging sonar in the aspect of target discrimination under the condition of reverberation, and improves the image visualization level of the high-frequency imaging sonar.
According to the content, the effective suppression of the interference image bright spots caused by reverberation in the original image of the high-frequency imaging sonar can be realized according to the following specific steps:
(1) analyzing the original image of the high-frequency imaging sonar by using an image connected domain analysis algorithm to determine an original imageThe number n of targets in the image, each target image is composed of a series of intrinsic pixel points, so that the intrinsic pixel point set A corresponding to each target image can be obtained through connected domain analysisi(i ═ 1,2,. n), where a isiAnd representing the set of the intrinsic pixel points corresponding to the ith target image.
(2) Based on the corresponding intrinsic pixel point set A of each target imageiSolving the generalized center e of each target imageiWherein e isiDenotes the generalized center of the ith target image, i 1, 2.
In the invention, the generalized center of the target image is defined as a plane expression of a three-dimensional space where the target image is located, and the form of the plane expression is as follows: ax + by + cz is 1, and the set A of the corresponding intrinsic pixel points of each target image is based oniThe space position of each pixel point in the image can be solved by using a least square method to solve the coefficients a, b and c of the mathematical expression, and then the generalized center e of the ith target image can be obtainedi1,2, a. Generalized center eiI.e. a least squares estimate of the mathematical expression describing the sample pixel points in the ith target image.
In the foregoing step (1), all target images in the original image are obtained through connected component analysis, and are recorded as a target image set
Without the original image being included in the set of target images
The residual pixel points are marked as the intrinsic pixel points of the non-target image
In the part of the non-target image, there may be an intrinsic pixel point that should be classified into the target image, so that the pixel points need to be identified and classified.
(3) Aiming at all the non-target image intrinsic pixel points in the original image
To n generalized centers e of the target imagei(i ═ 1, 2.. times, n), and the pixel point is drawn into the corresponding set of target image native pixel points on the minimum distance principle. Taking the pixel point j as an example, the distances from the pixel point j to the generalized centers of the 1 st, 2 nd and 3 … … n th target images can be calculated to be l respectively1、l2、l3、……、lnIf the smallest of the n distances is l2Then, the pixel point j is drawn into the intrinsic pixel point set A of the 2 nd target image2In (1). According to the method, the
Dividing each pixel point into an intrinsic pixel point set of the target image corresponding to the generalized center with the nearest distance to obtain an extended pixel point set A 'of the target image'i. When the point is on
After each pixel point in the target image is divided, an extended pixel point set A 'of each target image is obtained'i(i ═ 1,2,. n), satisfy
Thereby completing the classification of the non-target intrinsic pixel points.
(4) For each extended pixel point set A'i(i ═ 1, 2.. multidot.n), calculate a'iTo the generalized center e of the target imageiThe distance L of (a) is set, and a probability distribution set D of discrete random variables approximately satisfying a normal distribution (or a one-sided normal distribution) is formed by using the distance L as a set elementi(i＝1,2,...,n)。
(5) For each set DiCalculating the standard deviation (mean square error) σ of all elements in the seti(σiNot less than 0) to each DiSet significance level αiThe confidence level can be found from the standard normal distribution probability density function integral value table to be 1- αiUpper boundary value u of the samplei＝Φ-1(1-αi/2)(uiNot less than 0) to obtain DiCorresponding confidence levels of 1- αiWith a confidence interval of Ti＝[ei-uiσi,ei+uiσi]。
(6) For each set DiEach element in (1) is judged whether it falls within the confidence interval TiIn the method, the pixel points exceeding the confidence interval can be regarded as interference pixel points, and the pixel points falling in the confidence interval T are eliminatediPixel points corresponding to the other elements, namely the elimination extension pixel point set A'iMiddle to generalized center eiFall within the confidence interval TiThe outer sample pixel points. Thus, in the extended pixel point set A 'of the target image'iOn the basis of elements (sample pixel points) in (i ═ 1, 2., n), a new subset is constructed by eliminating partial pixel points with high probability as reverberation bright points of the underwater sound image
So that
To the generalized center e of the target imageiAll displacements of (2) fall within a confidence interval TiAnd finally, filtering the target underwater sound image reverberation bright spots based on normal distribution interval estimation.
(7) Record the final set of eigen pixel points for all target images as
This set represents all target images after reverberation suppression. Will be provided with
All the elements (sample pixel points) in the image are drawn on a display interface, namely the target underwater sound image with reverberation suppression is obtained.
In order to make the present invention more understandable to those skilled in the art, the above-described method is applied to a specific embodiment to show its technical effects.
As shown in fig. 1, the original display image obtained by the phased array three-dimensional acoustic imaging sonar detection in this embodiment has n-1 seabed target images, but obviously has more interference bright spots outside the target images, and it is necessary to perform reverberation suppression on the target images.
Reverberation suppression is performed on the original display image through the above steps (1) to (7), and two different significance levels α are set in the processi(0.5 and 0.32) controlling the pixel points of the original image to fall within the confidence interval of the corresponding target image by image analysis and processing, at αiIn the case of 0.5, the target underwater sound image after reverberation suppression is obtained as shown in fig. 2 at αiIn the case of 0.32, the target underwater sound image after reverberation suppression is obtained as shown in fig. 3.
Compared with the image processing effect of the submarine target, the underwater acoustic image reverberation suppression method based on the normal distribution interval estimation can obviously filter discrete and random reverberation bright spots in the original image, improve the application efficiency of imaging sonar and improve the visualization level of the target image.
The above-described embodiments are merely preferred embodiments of the present invention, which should not be construed as limiting the invention. Various changes and modifications may be made by one of ordinary skill in the pertinent art without departing from the spirit and scope of the present invention. Therefore, the technical scheme obtained by adopting the mode of equivalent replacement or equivalent transformation is within the protection scope of the invention.
Claims (6)
1. A normal distribution interval estimation-based underwater acoustic image reverberation suppression method is characterized by comprising the following steps:
s1: carrying out image connected domain analysis algorithm analysis on an original image of the high-frequency imaging sonar, and determining the number n of targets in the original image and an intrinsic pixel point set A corresponding to each target imagei，i＝1,2,...,n；
S2: based on books corresponding to each target imageSet of symptom pixel points AiSolving the generalized center e of each target imagei，i＝1,2,...,n；
S3: aiming at all the non-target image intrinsic pixel points in the original image, calculating the generalized center e of each pixel point to n target imagesiAnd dividing the target image into the intrinsic pixel point set of the target image corresponding to the nearest generalized center, so that each target image obtains the extended pixel point set A'iFinishing the classification of the intrinsic pixel points of the non-target image;
s4: for each extended pixel point set A'iCalculating each pixel point in the set to the generalized center e of the target imageiAnd constructing a set D with the distance as an elementi；
S5: for each set DiPerforming confidence interval estimation on the elements in the normal distribution, and removing an extended pixel point set A'iAnd (4) filtering the reverberation bright points of the underwater sound image estimated based on the normal distribution interval by taking the removed extended pixel point set as a final intrinsic pixel point set corresponding to the target image according to the sample pixel points of which the distance from the mid-to-generalized center is outside the confidence interval.
2. The method as claimed in claim 1, wherein the generalized central form is a mathematical expression ax + by + cz-1 of a three-dimensional space where the target image is located, and a least square method is used to solve a coefficient of the mathematical expression, that is, the target image a is obtainediGeneralized center e ofiGeneralized center eiA least squares estimate representing a mathematical expression describing sample pixel points of a target image.
3. The method as claimed in claim 1, wherein the normal distribution interval estimation-based reverberation suppression method for the underwater acoustic image is characterized in that the non-target image intrinsic pixel points are
Representing the set of all but target images in the original image
And the other pixel points are not.
4. The method as claimed in claim 1, wherein the set D is the set of underwater acoustic image reverberation suppression methods based on normal distribution interval estimationiApproximately satisfying a normal distribution or a unilateral normal distribution.
5. The method for suppressing reverberation of underwater sound image based on normal distribution interval estimation as claimed in claim 1, wherein in S5, aiming at set DiAll elements in (2) are calculated for their standard deviation σiAnd set significance level αiTo a confidence level of 1- αiConfidence interval of, then from A'iEliminating pixel points corresponding to elements outside the confidence interval to obtain a new subset
As the final set of native pixel points for the corresponding target image.
6. The method as claimed in claim 5, wherein the final eigen pixel point set of all target images is selected as the basis for the reverberation suppression of the underwater acoustic image based on the normal distribution interval estimation
All the elements in (2) are drawn on a display interface to serve as a target underwater sound image for reverberation suppression.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010139790.7A CN111325696B (en) | 2020-03-03 | 2020-03-03 | Underwater sound image reverberation suppression method based on normal distribution interval estimation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010139790.7A CN111325696B (en) | 2020-03-03 | 2020-03-03 | Underwater sound image reverberation suppression method based on normal distribution interval estimation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111325696A true CN111325696A (en) | 2020-06-23 |
| CN111325696B CN111325696B (en) | 2023-07-07 |
Family
ID=71167269
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010139790.7A Active CN111325696B (en) | 2020-03-03 | 2020-03-03 | Underwater sound image reverberation suppression method based on normal distribution interval estimation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111325696B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114462458A (en) * | 2022-04-11 | 2022-05-10 | 自然资源部第一海洋研究所 | Ship underwater signal noise reduction and target enhancement method |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102930273A (en) * | 2012-10-15 | 2013-02-13 | 西安电子科技大学 | Auroral oval segmenting method based on brightness self-adaptive level set |
| CN103714542A (en) * | 2013-12-25 | 2014-04-09 | 中国人民解放军海军工程大学 | Extraction method for target highlight in low-resolution high-frequency sonar image |
| US20140133750A1 (en) * | 2012-11-12 | 2014-05-15 | FMV Innovations, LLC | Method for improving images captured underwater |
| US20140238136A1 (en) * | 2011-09-26 | 2014-08-28 | Ontario Power Generation Inc. | Ultrasound matrix inspection |
| CN104880703A (en) * | 2014-02-27 | 2015-09-02 | 中国科学院声学研究所 | Side-scan sonar target detection reverberation suppressing technologies |
| CN105741284A (en) * | 2016-01-28 | 2016-07-06 | 中国船舶重工集团公司第七一〇研究所 | Multi-beam forward-looking sonar target detection method |
| US20190089869A1 (en) * | 2017-09-21 | 2019-03-21 | United States Of America As Represented By Secretary Of The Navy | Single Image Haze Removal |
| US20190114747A1 (en) * | 2016-04-07 | 2019-04-18 | Carmel Haifa University Economic Corporation Ltd. | Image dehazing and restoration |
| CN110852959A (en) * | 2019-10-14 | 2020-02-28 | 江苏帝一集团有限公司 | Sonar image filtering method based on novel median filtering algorithm |
| CN111462729A (en) * | 2020-03-31 | 2020-07-28 | 因诺微科技(天津)有限公司 | Fast language identification method based on phoneme log-likelihood ratio and sparse representation |
-
2020
- 2020-03-03 CN CN202010139790.7A patent/CN111325696B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140238136A1 (en) * | 2011-09-26 | 2014-08-28 | Ontario Power Generation Inc. | Ultrasound matrix inspection |
| CN102930273A (en) * | 2012-10-15 | 2013-02-13 | 西安电子科技大学 | Auroral oval segmenting method based on brightness self-adaptive level set |
| US20140133750A1 (en) * | 2012-11-12 | 2014-05-15 | FMV Innovations, LLC | Method for improving images captured underwater |
| CN103714542A (en) * | 2013-12-25 | 2014-04-09 | 中国人民解放军海军工程大学 | Extraction method for target highlight in low-resolution high-frequency sonar image |
| CN104880703A (en) * | 2014-02-27 | 2015-09-02 | 中国科学院声学研究所 | Side-scan sonar target detection reverberation suppressing technologies |
| CN105741284A (en) * | 2016-01-28 | 2016-07-06 | 中国船舶重工集团公司第七一〇研究所 | Multi-beam forward-looking sonar target detection method |
| US20190114747A1 (en) * | 2016-04-07 | 2019-04-18 | Carmel Haifa University Economic Corporation Ltd. | Image dehazing and restoration |
| US20190089869A1 (en) * | 2017-09-21 | 2019-03-21 | United States Of America As Represented By Secretary Of The Navy | Single Image Haze Removal |
| CN110852959A (en) * | 2019-10-14 | 2020-02-28 | 江苏帝一集团有限公司 | Sonar image filtering method based on novel median filtering algorithm |
| CN111462729A (en) * | 2020-03-31 | 2020-07-28 | 因诺微科技(天津)有限公司 | Fast language identification method based on phoneme log-likelihood ratio and sparse representation |
Non-Patent Citations (4)
| Title |
|---|
| KAY L GEMBA, ET.AL: "Source characterization using recordings made in a reverberant underwater channel" * |
| MELITA A. SAMOILYS, ET.AL: "Determining methods of underwater visual census for estimating the abundance of coral reef fishes" * |
| 宋泽林: "基于时频域滤波的亮点特征提取及识别研究" * |
| 王志杰等: "水下目标声图像特征分析与二次处理" * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114462458A (en) * | 2022-04-11 | 2022-05-10 | 自然资源部第一海洋研究所 | Ship underwater signal noise reduction and target enhancement method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111325696B (en) | 2023-07-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109360156B (en) | Single image rain removing method based on image block generation countermeasure network | |
| CN106204467B (en) | Image denoising method based on cascade residual error neural network | |
| CN102306377B (en) | Method and device for reducing noise in ultrasound image | |
| Mallawaarachchi et al. | Spectrogram denoising and automated extraction of the fundamental frequency variation of dolphin whistles | |
| CN109100788B (en) | Seismic data non-local mean de-noising method | |
| WO2008016109A1 (en) | Learning data set optimization method for signal identification device and signal identification device capable of optimizing the learning data set | |
| CN106651813B (en) | A kind of non-local mean denoising method of more fragments compositings | |
| CN110390646B (en) | Detail-preserving image denoising method | |
| CN111460953B (en) | Electrocardiosignal classification method based on self-adaptive learning of countermeasure domain | |
| Feng et al. | A new edge detection algorithm based on Canny idea | |
| CN104915930B (en) | The method and device of grey level compensation and noise suppressed is carried out to image | |
| CN102663706A (en) | Adaptive weighted mean value filtering method based on diamond template | |
| CN111505651B (en) | Feature extraction method for potential moving object of active sonar echo map | |
| Evans et al. | Mode filtering to reduce ultrasound speckle for feature extraction | |
| CN111325696A (en) | Underwater acoustic image reverberation suppression method based on normal distribution interval estimation | |
| CN104200434B (en) | Non-local mean image denoising method based on noise variance estimation | |
| Yin et al. | Sonar image target detection based on deep learning | |
| CN107610056B (en) | Mixed weighting wiener filtering image denoising method based on total variation | |
| Han et al. | A novel sonar image denoising algorithm based on block matching | |
| CN114926377B (en) | Modeling method of side-scan sonar gain compensation judgment model | |
| Karthiha et al. | Speckle Noise Suppression in Ultrasound Images Using Modular Neural Networks. | |
| CN109427042B (en) | Method for extracting layered structure and spatial distribution of local sea area sedimentary layer | |
| Enríquez et al. | Image denoising using block matching and discrete cosine transform with edge restoring | |
| LU503334B1 (en) | Forward-looking sonar target recognition method based on asynchronous learning factors | |
| CN117115452A (en) | Controllable medical ultrasonic image denoising method, system and computer storage medium |
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 |