CN116311212A - Ship number identification method and device based on high-speed camera and in motion state - Google Patents
Ship number identification method and device based on high-speed camera and in motion state Download PDFInfo
- Publication number
- CN116311212A CN116311212A CN202310537902.8A CN202310537902A CN116311212A CN 116311212 A CN116311212 A CN 116311212A CN 202310537902 A CN202310537902 A CN 202310537902A CN 116311212 A CN116311212 A CN 116311212A
- Authority
- CN
- China
- Prior art keywords
- image
- region
- extremum
- ship
- motion state
- 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 58
- 238000012545 processing Methods 0.000 claims abstract description 53
- 238000012937 correction Methods 0.000 claims abstract description 22
- 238000010606 normalization Methods 0.000 claims abstract description 11
- 238000003708 edge detection Methods 0.000 claims description 18
- 239000013598 vector Substances 0.000 claims description 18
- 238000004458 analytical method Methods 0.000 claims description 17
- 230000009466 transformation Effects 0.000 claims description 13
- 230000008569 process Effects 0.000 claims description 12
- 238000009499 grossing Methods 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 5
- 230000004927 fusion Effects 0.000 claims description 5
- 230000001629 suppression Effects 0.000 claims description 5
- 238000003702 image correction Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 230000006870 function Effects 0.000 description 26
- 238000004891 communication Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 8
- 238000007726 management method Methods 0.000 description 7
- 238000013473 artificial intelligence Methods 0.000 description 4
- 238000004590 computer program Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000035945 sensitivity Effects 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 235000002566 Capsicum Nutrition 0.000 description 1
- 239000006002 Pepper Substances 0.000 description 1
- 235000016761 Piper aduncum Nutrition 0.000 description 1
- 235000017804 Piper guineense Nutrition 0.000 description 1
- 244000203593 Piper nigrum Species 0.000 description 1
- 235000008184 Piper nigrum Nutrition 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011946 reduction process Methods 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
- 238000011426 transformation method Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/243—Aligning, centring, orientation detection or correction of the image by compensating for image skew or non-uniform image deformations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
Abstract
The invention relates to the technical field of ship number identification, and discloses a ship number identification method and device based on a high-speed camera in a motion state, wherein the method comprises the following steps: acquiring a ship image of a ship in a motion state, performing image graying treatment on the ship image to obtain a graying image, performing histogram equalization treatment on the graying image to obtain an equalized image, and performing inclination correction treatment on a denoising image to obtain a corrected image; calculating extremum regions of the corrected image based on a preset extremum region algorithm, identifying region categories of each region in the extremum regions, and extracting character regions in the classified extremum regions to obtain target character regions; and carrying out normalization processing on the characters in the target character area to obtain normalized characters, and inputting the normalized characters into a pre-trained character recognition model to recognize the characters in the target character area, so as to obtain the target ship number. The invention aims to improve the accuracy of identifying the ship number in a motion state.
Description
Technical Field
The invention relates to the technical field of ship number identification, in particular to a ship number identification method and device based on a high-speed camera under a motion state.
Background
The ship is a core carrier for waterway transportation, and the key point of managing water area traffic is to perfect the management of ship identity information, so that the ship number identification is an application of a high-speed camera image identification technology in ship license plate identification, and the ship number identification technology is required to extract and identify the ship number in a motion state from complex scenes and is widely applied to water area traffic supervision.
At present, ship number identification is mainly based on equipment such as satellite remote sensing images, radars and unmanned aerial vehicle shooting to acquire ship images, and ship number identification is realized by utilizing the ais (ship automatic identification device) technology, but in an actual service scene, a ship body is easily interfered by wind, sun, light, dust, water stains and the like, so that the phenomenon of image blurring occurs when the equipment such as satellite remote sensing images, radars and unmanned aerial vehicle shooting is utilized to acquire ship images, the accuracy of subsequent ship number identification is further influenced, and under complex scenes and under a ship motion state, the ship number identification is difficult due to the fact that the ship number is blurring and is blocked. Therefore, the accuracy of the identification of the ship number in the moving state is insufficient.
Disclosure of Invention
The invention provides a ship number identification method and device based on a high-speed camera in a motion state, and mainly aims to improve the accuracy of ship number identification in the motion state.
In order to achieve the above object, the invention provides a ship number identification method based on a high-speed camera in a motion state, comprising the following steps:
acquiring a ship image of a ship in a motion state by using a high-speed camera, performing image graying treatment on the ship image to obtain a graying image, performing histogram equalization treatment on the graying image to obtain an equalized image, performing smoothing denoising treatment on the equalized image to obtain a denoised image, and performing inclination correction treatment on the denoised image to obtain a corrected image;
calculating an extremum region of the corrected image based on a preset extremum region algorithm, identifying a region category of each region in the extremum region, classifying the extremum region according to the region category to obtain a classified extremum region, and extracting a character region in the classified extremum region to obtain a target character region;
and carrying out normalization processing on the characters in the target character area to obtain normalized characters, and inputting the normalized characters into a pre-trained character recognition model to recognize the characters in the target character area to obtain a target ship number.
Optionally, the performing image graying processing on the ship image to obtain a graying image includes:
performing image graying treatment on the ship image by the following formula:
here, grab (m, n) represents a graying image obtained by performing image graying processing on a ship image, R, G, B represents the values of red, green and blue channels of each pixel point in the image, and m and n represent coordinates of the pixel point.
Optionally, performing histogram equalization processing on the grayscale image to obtain an equalized image, including:
performing wavelet transformation on the gray-scale image to obtain sub-band images with different frequencies;
performing histogram equalization processing on each sub-band image in the sub-band images to obtain equalized sub-band images;
performing wavelet inverse transformation on the balanced subband image to obtain a transformed image;
and carrying out image fusion on the transformed image to obtain an equalized image.
Optionally, the performing smoothing denoising processing on the equalized image to obtain a denoised image includes:
discretizing the equalization image to obtain a discretized image;
defining an objective function corresponding to the discretized image;
And solving an image noise value corresponding to the discretized image by using the objective function, wherein the objective function comprises:
wherein ,
representing the image noise value corresponding to said discretized image,>
gray value sum representing non-noise part in said discretized image,/>
A gray value sum representing a noise portion in the discretized image;
and when the image noise value is larger than a preset noise value, denoising the image noise corresponding to the discretized image to obtain a denoised image.
Optionally, the calculating the extremum region of the corrected image based on the preset extremum region algorithm includes:
detecting image pixel points of the corrected image;
defining a sliding window of the corrected image according to the image pixel points;
traversing the correction image according to the sliding window to obtain a traversed image;
calculating a pixel minimum value and a pixel maximum value in the traversal image;
if the pixel minimum value and the pixel maximum value are positioned at the center of the sliding window, recording window coordinates of the sliding window;
taking the corresponding region of the window coordinate in the corrected image as the extremum region of the corrected image
Optionally, the classifying the extremum area according to the area category to obtain a classified extremum area includes:
extracting a feature vector from each of the extremum regions;
clustering the feature vectors by using a preset clustering algorithm to obtain clustered vectors;
performing index analysis on the clustering vectors to obtain analysis results;
calculating a deviation value corresponding to the analysis result according to the analysis result;
if the deviation value is not greater than a preset threshold value, adjusting algorithm parameters of the clustering algorithm to obtain a target clustering algorithm;
and classifying the extremum regions by using the target clustering algorithm to obtain classified extremum regions.
Optionally, the extracting the character region in the classifying extremum region to obtain a target character region includes:
detecting edges in the classifying extremum areas by using a preset edge detection algorithm to obtain detected edges;
performing enhancement treatment on the detection edge to obtain an enhancement edge;
performing connection treatment on the reinforced edge to obtain a connection edge;
detecting an edge shape and a text area corresponding to the connecting edge;
and combining the edge shapes to determine a target character area in the text area.
Optionally, the detecting the edge in the classification extremum area by using a preset edge detection algorithm to obtain a detected edge includes:
calculating a region gradient value in the classifying extremum region by using the preset edge detection algorithm, wherein the preset edge detection algorithm comprises the following steps:
wherein G represents the regional gradient value,
representing gradient values in horizontal direction in the region of the classification extremum,/->
Representing gradient values in the vertical direction in the classification extremum region, g representing an edge detection operator, and A representing the classification extremum region;
according to the regional gradient values, performing non-maximum suppression on the gradient image corresponding to the regional gradient values to obtain a target gradient image;
performing double-threshold detection on the target gradient image to obtain edge intensity and edge quantity corresponding to the target gradient image;
and detecting edges in the classifying extremum areas according to the edge intensity and the edge quantity to obtain detected edges.
Optionally, the determining, in combination with the edge shape, a target character area in the text area includes:
performing region segmentation on the text region to obtain a segmented region;
Identifying region characters in the partitioned region;
carrying out semantic analysis on the regional characters to obtain character semantics;
determining a first character area in the partitioned areas according to the character semantics;
filtering the first character area to obtain a second character area;
and determining a target character area in the text area from the second character area according to the edge shape.
In order to solve the above problems, the present invention also provides a ship number recognition device based on a high-speed camera in a motion state, the device comprising:
the image correction module is used for acquiring ship images of the ship in a motion state by using a high-speed camera, carrying out image graying treatment on the ship images to obtain graying images, carrying out histogram equalization treatment on the graying images to obtain equalized images, carrying out smooth denoising treatment on the equalized images to obtain denoised images, and carrying out inclination correction treatment on the denoised images to obtain corrected images;
the region extraction module is used for calculating an extremum region of the corrected image based on a preset extremum region algorithm, identifying a region category of each region in the extremum region, classifying the extremum region according to the region category to obtain a classified extremum region, and extracting a character region in the classified extremum region to obtain a target character region;
The ship number recognition module is used for carrying out normalization processing on the characters in the target character area to obtain normalized characters, and inputting the normalized characters into a pre-trained character recognition model to recognize the characters in the target character area to obtain the target ship number.
According to the invention, firstly, a ship image of the ship in a motion state is acquired by utilizing a high-speed camera, so that the ship image of the ship in the motion state can be conveniently processed, further, the embodiment of the invention can obtain a grey image by carrying out image grey treatment on the ship image, because the character content in the ship image is irrelevant to color information and the original color image occupies a larger storage, the processing speed of the whole algorithm can be improved, further, the embodiment of the invention can obtain an equalized image by carrying out histogram equalization treatment on the grey image, the main effect of the histogram equalization is that a target with unobvious recognition target is influenced by light, further, the invention can obtain a de-noised image by carrying out smooth de-noising treatment on the equalized image, the noise point in the image can be removed as much as possible, the subsequent character recognition is prevented from having a larger error, the invention can carry out inclination correction treatment on the de-noised image, can reduce the processing speed of the whole algorithm, further, the invention can be used for correcting the extreme value region of the character is based on the subsequent character region, the extreme value region is corrected by carrying out the correction on the extreme value region of the image, the extreme value region is based on the character region is corrected by carrying out the correction on the extreme value region in the image, the invention, the extreme value region is corrected by carrying out the correction of the extreme value region is based on the subsequent region of the character region is corrected, the character set can be processed so as to facilitate the recognition of a subsequent model, and the normalized characters are input into a pre-trained character recognition model so as to recognize the characters of the target character area, so that the ship number of the character area is recognized. Therefore, the invention provides a method and a device for identifying the ship number under the motion state based on a high-speed camera, and the method and the device mainly aim to improve the accuracy of identifying the ship number under the motion state.
Drawings
Fig. 1 is a flow chart of a method for identifying a ship number based on a high-speed camera in a motion state according to an embodiment of the present invention;
fig. 2 is a functional block diagram of a ship number recognition device in a motion state based on a high-speed camera according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an electronic device for implementing the ship number recognition method based on the high-speed camera in a motion state according to an embodiment of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
The embodiment of the application provides a ship number identification method based on a high-speed camera in a motion state. In the embodiment of the present application, the execution body of the method for identifying a ship number based on the high-speed camera in a motion state includes, but is not limited to, at least one of a server, a terminal, and an electronic device capable of being configured to execute the method provided in the embodiment of the present application. In other words, the method for identifying the ship number based on the high-speed camera in the motion state can be executed by software or hardware installed in a terminal device or a server device, wherein the software can be a blockchain platform. The service end includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like. The server may be an independent server, or may be a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms.
Referring to fig. 1, a flow chart of a method for identifying a ship number based on a high-speed camera in a moving state according to an embodiment of the present invention is shown. In this embodiment, the method for identifying the ship number based on the high-speed camera in the motion state includes steps S1 to S3:
s1, acquiring ship images of a ship in a motion state by using a high-speed camera, performing image graying treatment on the ship images to obtain gray-scale images, performing histogram equalization treatment on the gray-scale images to obtain equalized images, performing smooth denoising treatment on the equalized images to obtain denoising images, and performing inclination correction treatment on the denoising images to obtain corrected images.
According to the invention, the ship image of the ship in the motion state is acquired by using the high-speed camera, so that the ship image of the ship in the motion state can be obtained, and the subsequent image processing can be conveniently carried out on the ship image. Wherein the high-speed camera is a device capable of capturing a moving image at an exposure of less than 1/1000 seconds or a frame rate exceeding 250 frames per second, is used for recording a fast moving object as a photo image onto a storage medium, has characteristics of high resolution, high sensitivity, high frame rate, fast exposure time, and the like, and the ship image is a photo or image reflecting ship appearance information collected by the high-speed camera, and is used for identification and tracking of a ship, and safety management of the ship. Optionally, the ship image may be captured by using a high-speed shutter and a high-speed light sensor in the high-speed video camera, where the high-speed shutter and the high-speed light sensor are all technologies commonly used in the field of photography, and are used for shooting an object moving at a high speed or shooting a clear picture in a low-light environment, the high-speed shutter refers to a shooting mode in which the shutter speed of the camera is very fast, the faster the shutter speed, the camera can capture a motion track of the object in a shorter time, so that a clearer picture is shot, the higher the sensitivity of the sensor is a technology in which the sensitivity of the camera sensor is very high, the camera can shoot a clearer picture in a low-light environment, and the high-speed shutter and the high-speed light sensor generally need to be used in a matching manner, so as to obtain the best shooting effect in different scenes.
Further, in the embodiment of the invention, the ship image is subjected to image graying processing to obtain the graying image, and the character content in the ship image is irrelevant to the color information, and the original color image occupies more storage, so that the processing speed of the whole algorithm can be improved. Wherein the graying image refers to an image without color information, each pixel of which is described by one quantized gray scale. Alternatively, the image graying process of the ship image is a process of converting a color image into a gray image, so that the complexity of the image processing can be simplified while main information of the image is maintained, and the graying process can be realized by a weighted average method. Wherein, the weighted average method refers to the importance and other indexes.
As one embodiment of the present invention, the performing image graying processing on the ship image to obtain a grayed image includes: performing image graying treatment on the ship image by the following formula:
here, grab (m, n) represents a graying image obtained by performing image graying processing on a ship image, R, G, B represents the values of red, green and blue channels of each pixel point in the image, and m and n represent coordinates of the pixel point.
Further, in the embodiment of the invention, the histogram equalization processing is performed on the gray-scale image to obtain an equalized image, and the main function of the histogram equalization is to highlight the target which is affected by light and causes the identification target to be unobvious. The equalized image refers to an image with increased image gray scale interval, increased image contrast and approximately the same pixel point for each gray scale.
As one embodiment of the present invention, performing histogram equalization processing on the gray-scaled image to obtain an equalized image, including: performing wavelet transformation on the gray-scale image to obtain sub-band images with different frequencies; performing histogram equalization processing on each sub-band image in the sub-band images to obtain equalized sub-band images; performing wavelet inverse transformation on the balanced subband image to obtain a transformed image; and carrying out image fusion on the transformed image to obtain an equalized image.
The sub-band image is an image obtained by decomposing an original image into sub-signals or sub-images with different frequencies through the wavelet transformation, the equalizing sub-band image is an image obtained by carrying out histogram equalization processing on each sub-band image, and the transformation image is an image obtained by carrying out wavelet inverse transformation on the equalizing sub-band image.
Further, the wavelet transformation is a mathematical transformation method capable of decomposing a signal or an image into sub-signals or sub-images with different frequencies, and comprises a low-frequency sub-band image and a high-frequency sub-band image, wherein the low-frequency sub-band image comprises most of energy of an original image, the high-frequency sub-band image comprises detail information of the image, the wavelet inverse transformation is a process of recovering coefficients after the wavelet transformation into the original signal, the basic idea of the wavelet inverse transformation is that wavelet coefficients are gradually reconstructed into the original signal through an inverse filter and an inverse downsampling operation, the histogram equalization processing can be realized through a histogram equalization algorithm based on frequency division and fusion, the histogram equalization algorithm based on frequency division and fusion is that the histogram equalization processing is performed on the low-frequency component in consideration of separating the high-frequency component from the low-frequency component of the image, the high-frequency component is subjected to linear weighting enhancement, and then the two are fused, so that the problems of losing of detail information and noise amplification of the image caused by the histogram equalization algorithm can be avoided.
Further, the method and the device obtain the denoising image by carrying out smoothing denoising treatment on the equalization image, and the smoothing denoising treatment can remove noise points in the image as much as possible, so that larger errors in subsequent character recognition are avoided. The denoising image refers to an image with noise points eliminated. Alternatively, the smoothing noise reduction process may be implemented by a total variation denoising method, in which an energy function of an image is constructed and minimized so that the image reaches a smooth state.
As an embodiment of the present invention, the performing a smoothing denoising process on the equalized image to obtain a denoised image includes: discretizing the equalization image to obtain a discretized image; defining an objective function corresponding to the discretized image; and solving an image noise value corresponding to the discretized image by using the objective function, wherein the objective function comprises:
wherein ,
representation houseImage noise value corresponding to the discretized image, < >>
Gray value sum representing non-noise part in said discretized image,/>
A gray value sum representing a noise portion in the discretized image;
and when the image noise value is larger than a preset noise value, denoising the image noise corresponding to the discretized image to obtain a denoised image.
The discretized image is an image obtained by discretizing the original image, the objective function is a function of calculated noise corresponding to the discretized image, and the image noise is noise existing in the noise reduction image, such as salt and pepper noise.
Optionally, the discretization refers to a process of converting continuous image data into discrete image data, the purpose of the image discretization is to convert continuous gray values into discrete gray levels, so as to facilitate digital image processing and analysis, an objective function corresponding to the definition of the discretized image can be implemented through an embedded function, and denoising processing on the image noise can be implemented through a denoising function, and the denoising function is compiled by a scripting language.
According to the invention, through carrying out inclination correction processing on the denoising images, the angles of some images in the denoising images can be corrected, so that the difficulty of subsequent image processing is reduced, wherein the corrected images are images with the inclination angles adjusted. Alternatively, the tilt correction process may be implemented by a hough transform, which is a very important method of detecting the boundary shape of the break point, by transforming the image coordinate space into the parameter space to achieve the fitting of the straight line and the curve.
S2, calculating an extremum region of the corrected image based on a preset extremum region algorithm, identifying a region category of each region in the extremum region, classifying the extremum region according to the region category to obtain a classified extremum region, and extracting a character region in the classified extremum region to obtain a target character region.
According to the method, the extremum regions of the corrected image are calculated based on a preset extremum region algorithm, so that different extremum regions can be obtained, and the extremum regions can be classified later. The extreme value region algorithm is an algorithm for searching extreme value points in an image, and further, the extreme value region algorithm is thought to search local maximum values and local minimum values in the image so as to determine an extreme value region, wherein the extreme value region refers to a local region with the maximum value or the minimum value in the image, and in digital image processing, the extreme value region is generally used in applications such as image segmentation, target detection, feature extraction and the like.
As an embodiment of the present invention, the calculating the extremum region of the corrected image based on the preset extremum region algorithm includes: detecting image pixel points of the corrected image; defining a sliding window of the corrected image according to the image pixel points; traversing the correction image according to the sliding window to obtain a traversed image; calculating a pixel minimum value and a pixel maximum value in the traversal image; if the pixel minimum value and the pixel maximum value are positioned at the center of the sliding window, recording window coordinates of the sliding window; and taking the corresponding region of the window coordinate in the corrected image as an extremum region of the corrected image.
Wherein the sliding window is a common technology in image processing, and can be used for processing local areas in an image, in the sliding window, the image is divided into a plurality of small blocks with equal size, each small block is called a window, and the traversing image is a graph obtained by traversing the correcting image by the sliding window.
Optionally, identifying the region class of each region in the extremum regions refers to dividing similar extremum regions into one class according to the size and position relationship between the extremum regions to obtain different region classes; and traversing the corrected image according to the sliding window, wherein the process of obtaining the traversed image is a process of performing sliding filtering on each pixel point in the corrected image by utilizing the sliding window.
According to the region category, the extremum regions are classified to obtain classified extremum regions, so that the classified extremum regions can be screened later.
As an embodiment of the present invention, classifying the extremum region according to the region category to obtain a classified extremum region includes: extracting a feature vector from each of the extremum regions; clustering the feature vectors by using a preset clustering algorithm to obtain clustered vectors; performing index analysis on the clustering vectors to obtain analysis results; calculating a deviation value corresponding to the analysis result according to the analysis result; if the deviation value is not greater than a preset threshold value, adjusting algorithm parameters of the clustering algorithm to obtain a target clustering algorithm; and classifying the extremum regions by using the target clustering algorithm to obtain classified extremum regions.
Wherein the feature vector is a vector for describing data features, and can be used for representing certain attributes or features of data, the feature vector can be used for describing the features of an extremum region in image processing, the clustering algorithm is a density-based algorithm, the algorithm identifies clusters by determining the density of data points, the data points with high density are clustered together to form clusters, the data points with low density are regarded as noise or abnormal points, the basic idea of the algorithm is that the data points are divided into three types of core points, boundary points and noise points, the core points are included in a radius with the point as a center, the points can form a cluster, the boundary points are included in a radius with the point as a center, the number of the included data points is insufficient to form a cluster, the points belong to the boundary of a certain cluster, the clustering vector is a vector obtained by clustering the feature vector by the preset clustering algorithm, and the deviation value is the deviation degree of the analysis result from the true value.
The character recognition method and the character recognition device are used for carrying out character recognition on the character areas by extracting the character areas in the classifying extremum areas. Wherein the character area generally refers to an area of the image that contains characters, which may be letters, numbers, symbols, or other special characters.
As an embodiment of the present invention, the extracting the character region in the classifying extremum region to obtain the target character region includes: detecting edges in the classifying extremum areas by using a preset edge detection algorithm to obtain detected edges; performing enhancement treatment on the detection edge to obtain an enhancement edge; performing connection treatment on the reinforced edge to obtain a connection edge; detecting an edge shape and a text area corresponding to the connecting edge; and combining the edge shapes to determine a target character area in the text area.
The detected edges are the places where the characteristics of color, brightness, texture and the like in the image are suddenly changed, and are usually caused by factors such as edges, shadows, illumination changes and the like of objects, and in digital image processing, the edges are very important characteristics because the edges can be used for identifying the objects, dividing the images, performing image enhancement and the like, and the enhancement processing is an image processing technology for enhancing edge information in the images so that the edges are clearer and more obvious.
Further, in an optional embodiment of the present invention, the detecting edges in the classification extremum area by using a preset edge detection algorithm to obtain detected edges includes: calculating a region gradient value in the classifying extremum region by using the preset edge detection algorithm, wherein the preset edge detection algorithm comprises the following steps:
wherein G represents the regional gradient value,
representing gradient values in horizontal direction in the region of the classification extremum,/->
Representing gradient values in the vertical direction in the classification extremum region, g representing an edge detection operator, and A representing the classification extremum region;
according to the regional gradient values, performing non-maximum suppression on the gradient image corresponding to the regional gradient values to obtain a target gradient image; performing double-threshold detection on the target gradient image to obtain edge intensity and edge quantity corresponding to the target gradient image; and detecting edges in the classifying extremum areas according to the edge intensity and the edge quantity to obtain detected edges.
The local gradient value refers to the direction and the change rate of the pixel value in the image, the non-maximum suppression is a common computer vision algorithm, and in the edge detection, the local maximum point is kept only in the edge detection result, and other points are removed, so that a more accurate edge detection result can be obtained, the target gradient image is an image obtained after the non-maximum suppression is performed on the gradient image, and the edge intensity represents the edge strength corresponding to the target gradient image.
Further, the dual-threshold detection is a commonly used image processing technique, which classifies pixels in an image into three categories: the specific implementation manner of the method is that pixel points in an image are divided into three sections below two thresholds, between the two thresholds and above the two thresholds according to gray values of the pixel points, and then the three sections are classified according to the sections where the pixel points are located, in the dual-threshold detection, the two thresholds are generally called a high threshold and a low threshold, the high threshold is used for identifying the strong edge, the low threshold is used for identifying the weak edge, and the ratio of the high threshold to the low threshold is generally between 2:1 and 3:1.
Further, as an optional embodiment of the present invention, the determining, in conjunction with the edge shape, a target character area in the text area includes: performing region segmentation on the text region to obtain a segmented region; identifying region characters in the partitioned region; carrying out semantic analysis on the regional characters to obtain character semantics; determining a first character area in the partitioned areas according to the character semantics; filtering the first character area to obtain a second character area; and determining a target character area in the text area from the second character area according to the edge shape.
The text region is divided into a text region and a text region, the text region is divided into a region corresponding to characters, the text region comprises character meanings corresponding to the character meanings, the first character region comprises the text meanings, and the second character region comprises the text region filtered by the first character region.
Further, the segmentation of the text region may be implemented by a segmenter, the recognition of the region characters in the segmented region may be implemented by an OCR word recognition technique, the semantic parsing of the region characters may be implemented by a semantic parser, and the filtering of the first character region may be implemented by a filter function, such as a filter function.
And S3, carrying out normalization processing on the characters in the target character area to obtain normalized characters, and inputting the normalized characters into a pre-trained character recognition model to recognize the characters in the target character area to obtain a target ship number.
The character of the target character area is normalized, so that the character can be processed in a character set to facilitate the recognition of a subsequent model, wherein the normalized character is obtained after the normalization of the character of the target character area, and further, the normalization of the character of the target character area can be realized through a linear normalization algorithm.
The normalized characters are input into a pre-trained character recognition model to recognize the characters of the target character area, so that the ship number of the character area can be recognized conveniently.
The character recognition model is a serial number recognition model, the structure of the character recognition model comprises a feature extraction layer and a serial number classification layer, the feature extraction layer can be composed of a convolution layer and a pooling layer, the serial number classification layer can be composed of a full connection layer and an activation function, the feature extraction layer is used for extracting features of the normalized characters, the serial number classification layer is used for linearly mapping the extracted features to serial number classes to obtain probability values of each serial number class, the serial number class corresponding to the maximum probability value is selected from the probability values as a target ship number, and the serial number classes refer to number classes which comprise ten number classes from 1 to 9.
According to the invention, firstly, a ship image of the ship in a motion state is acquired by utilizing a high-speed camera, so that the ship image of the ship in the motion state can be conveniently processed, further, the embodiment of the invention can obtain a grey image by carrying out image grey treatment on the ship image, because the character content in the ship image is irrelevant to color information and the original color image occupies a larger storage, the processing speed of the whole algorithm can be improved, further, the embodiment of the invention can obtain an equalized image by carrying out histogram equalization treatment on the grey image, the main effect of the histogram equalization is that a target with unobvious recognition target is influenced by light, further, the invention can obtain a de-noised image by carrying out smooth de-noising treatment on the equalized image, the noise point in the image can be removed as much as possible, the subsequent character recognition is prevented from having a larger error, the invention can carry out inclination correction treatment on the de-noised image, can reduce the processing speed of the whole algorithm, further, the invention can be used for correcting the extreme value region of the character is based on the subsequent character region, the extreme value region is corrected by carrying out the correction on the extreme value region of the image, the extreme value region is based on the character region is corrected by carrying out the correction on the extreme value region in the image, the invention, the extreme value region is corrected by carrying out the correction of the extreme value region is based on the subsequent region of the character region is corrected, the character set can be processed so as to facilitate the recognition of a subsequent model, and the normalized characters are input into a pre-trained character recognition model so as to recognize the characters of the target character area, so that the ship number of the character area is recognized. Therefore, the ship number identification method based on the high-speed camera in the embodiment of the invention can improve the accuracy of ship number identification in the motion state.
Fig. 2 is a functional block diagram of a ship number recognition device based on a high-speed camera in a motion state according to an embodiment of the present invention.
The ship number recognition device 100 based on the high-speed camera in the motion state can be installed in electronic equipment. The ship number recognition device 100 based on the high-speed camera in a motion state can comprise an image correction module 101, a region extraction module 102 and a ship number recognition module 103 according to the implemented functions. The module of the invention, which may also be referred to as a unit, refers to a series of computer program segments, which are stored in the memory of the electronic device, capable of being executed by the processor of the electronic device and of performing a fixed function.
In the present embodiment, the functions concerning the respective modules/units are as follows:
the image correction module 101 is configured to collect a ship image of a ship in a motion state by using a high-speed camera, perform image graying processing on the ship image to obtain a graying image, perform histogram equalization processing on the graying image to obtain an equalized image, perform smoothing denoising processing on the equalized image to obtain a denoised image, and perform inclination correction processing on the denoised image to obtain a corrected image;
The region extraction module 102 is configured to calculate an extremum region of the corrected image based on a preset extremum region algorithm, identify a region class of each region in the extremum region, classify the extremum region according to the region class, obtain a classified extremum region, and extract a character region in the classified extremum region to obtain a target character region;
the ship number recognition module 103 is configured to normalize the characters in the target character area to obtain normalized characters, and input the normalized characters into a pre-trained character recognition model to recognize the characters in the target character area, so as to obtain the target ship number.
In detail, each module in the device 100 for identifying a ship number in a moving state based on a high-speed camera in the embodiment of the present application adopts the same technical means as the method for identifying a ship number in a moving state based on a high-speed camera described in fig. 1, and can produce the same technical effects, which are not described herein again.
Fig. 3 is a schematic structural diagram of an electronic device 1 according to an embodiment of the present invention for implementing a ship number recognition method based on a high-speed camera in a motion state.
The electronic device 1 may comprise a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and may further comprise a computer program stored in the memory 11 and executable on the processor 10, such as a ship number identification method program in a motion state based on a high-speed camera.
The processor 10 may be formed by an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or may be formed by a plurality of integrated circuits packaged with the same function or different functions, including one or more central processing units (Central Processing Unit, CPU), a microprocessor, a digital processing chip, a graphics processor, a combination of various control chips, and so on. The processor 10 is a Control Unit (Control Unit) of the electronic device 1, connects respective parts of the entire electronic device using various interfaces and lines, executes or executes programs or modules stored in the memory 11 (for example, executes a ship number recognition method program or the like in a state of realizing movement based on a high-speed camera), and invokes data stored in the memory 11 to perform various functions of the electronic device and process the data.
The memory 11 includes at least one type of readable storage medium including flash memory, a removable hard disk, a multimedia card, a card type memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device, such as a mobile hard disk of the electronic device. The memory 11 may in other embodiments also be an external storage device of the electronic device, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the electronic device. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device. The memory 11 may be used not only for storing application software installed in an electronic device and various types of data, for example, codes of a ship number recognition method program based on a high-speed camera in a moving state, but also for temporarily storing data that has been output or is to be output.
The communication bus 12 may be a peripheral component interconnect standard (Peripheral Component Interconnect, PCI) bus, or an extended industry standard architecture (Extended Industry Standard Architecture, EISA) bus, among others. The bus may be classified as an address bus, a data bus, a control bus, etc. The bus is arranged to enable a connection communication between the memory 11 and at least one processor 10 etc.
The communication interface 13 is used for communication between the electronic device 1 and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and/or a wireless interface (e.g., WI-FI interface, bluetooth interface, etc.), typically used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a Display (Display), an input unit such as a Keyboard (Keyboard), or alternatively a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. The display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device and for displaying a visual user interface.
Fig. 3 shows only an electronic device with components, it being understood by a person skilled in the art that the structure shown in fig. 3 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than shown, or may combine certain components, or may be arranged in different components.
For example, although not shown, the electronic device 1 may further include a power source (such as a battery) for supplying power to each component, and preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that functions of charge management, discharge management, power consumption management, and the like are implemented through the power management device. The power supply may also include one or more of any of a direct current or alternating current power supply, recharging device, power failure detection circuit, power converter or inverter, power status indicator, etc. The electronic device 1 may further include various sensors, bluetooth modules, wi-Fi modules, etc., which will not be described herein.
It should be understood that the embodiments described are for illustrative purposes only and are not limited to this configuration in the scope of the patent application.
The program of the ship number recognition method stored in the memory 11 of the electronic device 1 and based on the high-speed camera in the motion state is a combination of a plurality of instructions, and when running in the processor 10, the method can be implemented:
acquiring a ship image of a ship in a motion state by using a high-speed camera, performing image graying treatment on the ship image to obtain a graying image, performing histogram equalization treatment on the graying image to obtain an equalized image, performing smoothing denoising treatment on the equalized image to obtain a denoised image, and performing inclination correction treatment on the denoised image to obtain a corrected image;
Calculating an extremum region of the corrected image based on a preset extremum region algorithm, identifying a region category of each region in the extremum region, classifying the extremum region according to the region category to obtain a classified extremum region, and extracting a character region in the classified extremum region to obtain a target character region;
and carrying out normalization processing on the characters in the target character area to obtain normalized characters, and inputting the normalized characters into a pre-trained character recognition model to recognize the characters in the target character area to obtain a target ship number.
In particular, the specific implementation method of the above instructions by the processor 10 may refer to the description of the relevant steps in the corresponding embodiment of the drawings, which is not repeated herein.
Further, the modules/units integrated in the electronic device 1 may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as separate products. The computer readable storage medium may be volatile or nonvolatile. For example, the computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM).
The present invention also provides a computer readable storage medium storing a computer program which, when executed by a processor of an electronic device, can implement:
acquiring a ship image of a ship in a motion state by using a high-speed camera, performing image graying treatment on the ship image to obtain a graying image, performing histogram equalization treatment on the graying image to obtain an equalized image, performing smoothing denoising treatment on the equalized image to obtain a denoised image, and performing inclination correction treatment on the denoised image to obtain a corrected image;
calculating an extremum region of the corrected image based on a preset extremum region algorithm, identifying a region category of each region in the extremum region, classifying the extremum region according to the region category to obtain a classified extremum region, and extracting a character region in the classified extremum region to obtain a target character region;
and carrying out normalization processing on the characters in the target character area to obtain normalized characters, and inputting the normalized characters into a pre-trained character recognition model to recognize the characters in the target character area to obtain a target ship number.
In the several embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be other manners of division when actually implemented.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
The embodiment of the application can acquire and process the related data based on the artificial intelligence technology. Among these, artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application that uses a digital computer or a digital computer-controlled machine to simulate, extend and expand human intelligence, sense the environment, acquire knowledge and use knowledge to obtain optimal results.
Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. The terms first, second, etc. are used to denote a name, but not any particular order.
Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.
Claims (10)
1. The ship number identification method based on the high-speed camera in the motion state is characterized by comprising the following steps of:
acquiring a ship image of a ship in a motion state by using a high-speed camera, performing image graying treatment on the ship image to obtain a graying image, performing histogram equalization treatment on the graying image to obtain an equalized image, performing smoothing denoising treatment on the equalized image to obtain a denoised image, and performing inclination correction treatment on the denoised image to obtain a corrected image;
calculating an extremum region of the corrected image based on a preset extremum region algorithm, identifying a region category of each region in the extremum region, classifying the extremum region according to the region category to obtain a classified extremum region, and extracting a character region in the classified extremum region to obtain a target character region;
and carrying out normalization processing on the characters in the target character area to obtain normalized characters, and inputting the normalized characters into a pre-trained character recognition model to recognize the characters in the target character area to obtain a target ship number.
2. The ship number identification method based on the high-speed camera in the motion state of claim 1, wherein the performing image graying processing on the ship image to obtain a graying image comprises:
Performing image graying treatment on the ship image by the following formula:
here, grab (m, n) represents a graying image obtained by performing image graying processing on a ship image, R, G, B represents the values of red, green and blue channels of each pixel point in the image, and m and n represent coordinates of the pixel point.
3. The ship number identification method based on the high-speed camera in the motion state of claim 1, wherein the histogram equalization processing is performed on the gray-scale image to obtain an equalized image, and the method comprises the following steps:
performing wavelet transformation on the gray-scale image to obtain sub-band images with different frequencies;
performing histogram equalization processing on each sub-band image in the sub-band images to obtain equalized sub-band images;
performing wavelet inverse transformation on the balanced subband image to obtain a transformed image;
and carrying out image fusion on the transformed image to obtain an equalized image.
4. The ship number identification method based on the high-speed camera in the motion state of claim 1, wherein the performing the smoothing denoising process on the equalized image to obtain a denoised image comprises:
discretizing the equalization image to obtain a discretized image;
Defining an objective function corresponding to the discretized image;
and solving an image noise value corresponding to the discretized image by using the objective function, wherein the objective function comprises:
wherein ,
representing the image noise value corresponding to said discretized image,>
gray value sum representing non-noise part in said discretized image,/>
A gray value sum representing a noise portion in the discretized image;
and when the image noise value is larger than a preset noise value, denoising the image noise corresponding to the discretized image to obtain a denoised image.
5. The ship number identification method based on the high-speed camera in the motion state according to claim 1, wherein the calculating the extremum area of the corrected image based on the preset extremum area algorithm comprises:
detecting image pixel points of the corrected image;
defining a sliding window of the corrected image according to the image pixel points;
traversing the correction image according to the sliding window to obtain a traversed image;
calculating a pixel minimum value and a pixel maximum value in the traversal image;
if the pixel minimum value and the pixel maximum value are positioned at the center of the sliding window, recording window coordinates of the sliding window;
And taking the corresponding region of the window coordinate in the corrected image as an extremum region of the corrected image.
6. The method for identifying a ship number in a motion state based on a high-speed camera according to claim 1, wherein the classifying the extremum area according to the area category to obtain a classified extremum area comprises:
extracting a feature vector from each of the extremum regions;
clustering the feature vectors by using a preset clustering algorithm to obtain clustered vectors;
performing index analysis on the clustering vectors to obtain analysis results;
calculating a deviation value corresponding to the analysis result according to the analysis result;
if the deviation value is not greater than a preset threshold value, adjusting algorithm parameters of the clustering algorithm to obtain a target clustering algorithm;
and classifying the extremum regions by using the target clustering algorithm to obtain classified extremum regions.
7. The method for identifying a ship number in a motion state based on a high-speed camera according to claim 1, wherein the extracting the character region in the classification extremum region to obtain a target character region comprises:
Detecting edges in the classifying extremum areas by using a preset edge detection algorithm to obtain detected edges;
performing enhancement treatment on the detection edge to obtain an enhancement edge;
performing connection treatment on the reinforced edge to obtain a connection edge;
detecting an edge shape and a text area corresponding to the connecting edge;
and combining the edge shapes to determine a target character area in the text area.
8. The method for identifying a ship number in a motion state based on a high-speed camera according to claim 7, wherein the detecting edges in the classification extremum area by using a preset edge detection algorithm to obtain detected edges comprises:
calculating a region gradient value in the classifying extremum region by using the preset edge detection algorithm, wherein the preset edge detection algorithm comprises the following steps:
wherein G represents the regional gradient value,
representing gradient values in horizontal direction in the region of the classification extremum,/->
Representing gradient values in the vertical direction in the classification extremum region, g representing an edge detection operator, and A representing the classification extremum region;
according to the regional gradient values, performing non-maximum suppression on the gradient image corresponding to the regional gradient values to obtain a target gradient image;
Performing double-threshold detection on the target gradient image to obtain edge intensity and edge quantity corresponding to the target gradient image;
and detecting edges in the classifying extremum areas according to the edge intensity and the edge quantity to obtain detected edges.
9. The method for identifying a ship number in a motion state based on a high-speed camera according to claim 7, wherein the determining a target character area in the text area in combination with the edge shape comprises:
performing region segmentation on the text region to obtain a segmented region;
identifying region characters in the partitioned region;
carrying out semantic analysis on the regional characters to obtain character semantics;
determining a first character area in the partitioned areas according to the character semantics;
filtering the first character area to obtain a second character area;
and determining a target character area in the text area from the second character area according to the edge shape.
10. The utility model provides a ship number recognition device based on high-speed camera realizes under motion state, its characterized in that, the device includes:
the image correction module is used for acquiring ship images of the ship in a motion state by using a high-speed camera, carrying out image graying treatment on the ship images to obtain graying images, carrying out histogram equalization treatment on the graying images to obtain equalized images, carrying out smooth denoising treatment on the equalized images to obtain denoised images, and carrying out inclination correction treatment on the denoised images to obtain corrected images;
The region extraction module is used for calculating an extremum region of the corrected image based on a preset extremum region algorithm, identifying a region category of each region in the extremum region, classifying the extremum region according to the region category to obtain a classified extremum region, and extracting a character region in the classified extremum region to obtain a target character region;
the ship number recognition module is used for carrying out normalization processing on the characters in the target character area to obtain normalized characters, and inputting the normalized characters into a pre-trained character recognition model to recognize the characters in the target character area to obtain the target ship number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310537902.8A CN116311212B (en) | 2023-05-15 | 2023-05-15 | Ship number identification method and device based on high-speed camera and in motion state |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310537902.8A CN116311212B (en) | 2023-05-15 | 2023-05-15 | Ship number identification method and device based on high-speed camera and in motion state |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116311212A true CN116311212A (en) | 2023-06-23 |
| CN116311212B CN116311212B (en) | 2023-08-18 |
Family
ID=86796253
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310537902.8A Active CN116311212B (en) | 2023-05-15 | 2023-05-15 | Ship number identification method and device based on high-speed camera and in motion state |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116311212B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101859382A (en) * | 2010-06-03 | 2010-10-13 | 复旦大学 | License plate detection and identification method based on maximum stable extremal region |
| CN107330365A (en) * | 2017-05-27 | 2017-11-07 | 深圳市美好幸福生活安全系统有限公司 | Traffic sign recognition method based on maximum stable extremal region and SVM |
| CN108416348A (en) * | 2018-01-29 | 2018-08-17 | 重庆邮电大学 | Plate location recognition method based on support vector machines and convolutional neural networks |
| CN110674812A (en) * | 2019-09-17 | 2020-01-10 | 沈阳建筑大学 | Civil license plate positioning and character segmentation method facing complex background |
| CN110738139A (en) * | 2019-09-26 | 2020-01-31 | 西安工程大学 | NIN license plate recognition method fusing Min-Max targets |
| US20210374466A1 (en) * | 2020-05-26 | 2021-12-02 | Zhejiang University | Water level monitoring method based on cluster partition and scale recognition |
| CN114092941A (en) * | 2020-08-03 | 2022-02-25 | 中移物联网有限公司 | Number identification method and device, electronic equipment and readable storage medium |
| CN114387591A (en) * | 2022-01-12 | 2022-04-22 | 平安普惠企业管理有限公司 | License plate recognition method, system, equipment and storage medium |
| CN114913518A (en) * | 2022-06-09 | 2022-08-16 | 平安科技（深圳）有限公司 | License plate recognition method, device, equipment and medium based on image processing |
-
2023
- 2023-05-15 CN CN202310537902.8A patent/CN116311212B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101859382A (en) * | 2010-06-03 | 2010-10-13 | 复旦大学 | License plate detection and identification method based on maximum stable extremal region |
| CN107330365A (en) * | 2017-05-27 | 2017-11-07 | 深圳市美好幸福生活安全系统有限公司 | Traffic sign recognition method based on maximum stable extremal region and SVM |
| CN108416348A (en) * | 2018-01-29 | 2018-08-17 | 重庆邮电大学 | Plate location recognition method based on support vector machines and convolutional neural networks |
| CN110674812A (en) * | 2019-09-17 | 2020-01-10 | 沈阳建筑大学 | Civil license plate positioning and character segmentation method facing complex background |
| CN110738139A (en) * | 2019-09-26 | 2020-01-31 | 西安工程大学 | NIN license plate recognition method fusing Min-Max targets |
| US20210374466A1 (en) * | 2020-05-26 | 2021-12-02 | Zhejiang University | Water level monitoring method based on cluster partition and scale recognition |
| CN114092941A (en) * | 2020-08-03 | 2022-02-25 | 中移物联网有限公司 | Number identification method and device, electronic equipment and readable storage medium |
| CN114387591A (en) * | 2022-01-12 | 2022-04-22 | 平安普惠企业管理有限公司 | License plate recognition method, system, equipment and storage medium |
| CN114913518A (en) * | 2022-06-09 | 2022-08-16 | 平安科技（深圳）有限公司 | License plate recognition method, device, equipment and medium based on image processing |
Non-Patent Citations (1)
| Title |
|---|
| 项?;李海鑫;张汝峰;褚俊英;郭宝军;乔东旭;: "结合边缘检测与横纵投影的车牌识别", 湖北农机化, no. 10 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116311212B (en) | 2023-08-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110956094B (en) | RGB-D multi-mode fusion personnel detection method based on asymmetric double-flow network | |
| CN108615226B (en) | Image defogging method based on generation type countermeasure network | |
| US9104914B1 (en) | Object detection with false positive filtering | |
| CN110298297B (en) | Flame identification method and device | |
| CN110929593B (en) | Real-time significance pedestrian detection method based on detail discrimination | |
| Pandey et al. | An efficient deep neural network with adaptive galactic swarm optimization for complex image text extraction | |
| CN109918971B (en) | Method and device for detecting number of people in monitoring video | |
| CN112686812B (en) | Bank card inclination correction detection method and device, readable storage medium and terminal | |
| CN111914698B (en) | Human body segmentation method, segmentation system, electronic equipment and storage medium in image | |
| CN110287791B (en) | Screening method and system for face pictures | |
| CN109255326B (en) | Traffic scene smoke intelligent detection method based on multi-dimensional information feature fusion | |
| KR101906796B1 (en) | Device and method for image analyzing based on deep learning | |
| CN110059634B (en) | Large-scene face snapshot method | |
| CN110807384A (en) | Small target detection method and system under low visibility | |
| CN112528917A (en) | Zebra crossing region identification method and device, electronic equipment and storage medium | |
| CN110969164A (en) | Low-illumination imaging license plate recognition method and device based on deep learning end-to-end | |
| Yan et al. | The recognition of traffic speed limit sign in hazy weather | |
| CN111027564A (en) | Low-illumination imaging license plate recognition method and device based on deep learning integration | |
| CN108537815B (en) | Video image foreground segmentation method and device | |
| CN112686248B (en) | Certificate increase and decrease type detection method and device, readable storage medium and terminal | |
| CN110910497B (en) | Method and system for realizing augmented reality map | |
| CN115147450B (en) | Moving target detection method and detection device based on motion frame difference image | |
| Pratomo et al. | Parking detection system using background subtraction and HSV color segmentation | |
| CN116311212B (en) | Ship number identification method and device based on high-speed camera and in motion state | |
| CN110633705A (en) | Low-illumination imaging license plate recognition method and device |
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 |