CN117078913B - Object inclination correction method, device, electronic equipment and storage medium - Google Patents
Object inclination correction method, device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN117078913B CN117078913B CN202311330100.6A CN202311330100A CN117078913B CN 117078913 B CN117078913 B CN 117078913B CN 202311330100 A CN202311330100 A CN 202311330100A CN 117078913 B CN117078913 B CN 117078913B
- Authority
- CN
- China
- Prior art keywords
- image
- signal
- preset
- tilt
- inclination
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012937 correction Methods 0.000 title claims abstract description 109
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000004458 analytical method Methods 0.000 claims abstract description 44
- 238000012545 processing Methods 0.000 claims abstract description 34
- 239000011159 matrix material Substances 0.000 claims description 44
- 230000009466 transformation Effects 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 18
- 238000010586 diagram Methods 0.000 description 10
- 238000013135 deep learning Methods 0.000 description 6
- 238000012549 training Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 5
- 238000012216 screening Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 3
- 238000002372 labelling Methods 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012015 optical character recognition Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000005311 autocorrelation function Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Classifications
-
- 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
-
- 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
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
- G06V10/507—Summing image-intensity values; Histogram projection analysis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the technical field of image processing, and provides an object inclination correction method, an object inclination correction device, electronic equipment and a storage medium, wherein the method comprises the following steps: acquiring a plurality of inclined images of a preset object; calculating a projection histogram of each inclined image in a preset direction, wherein the projection histogram of each inclined image comprises a plurality of columns of pixel points of each inclined image in the preset direction, and the value of each column is the number of effective pixel points belonging to a preset object on the column; taking the serial number of each column of the projection histogram of each inclined image as the moment and taking the value of each column as the signal value of the moment to obtain an inclined signal of each inclined image; performing signal analysis on the inclination signal of each inclination image to obtain the inclination signal intensity of each inclination signal; and determining a correction image meeting preset correction requirements from the plurality of inclined images according to the intensity of each inclined signal. Can efficiently achieve good correction effect.
Description
Technical Field
The present invention relates to the field of image processing technologies, and in particular, to an object inclination correction method, an object inclination correction device, an electronic device, and a storage medium.
Background
The inclination of the object in the image has great influence on the accurate identification of the object, and in order to improve the accuracy of the identification of the inclined object, one implementation way is to correct the inclined object in the image in an image processing mode according to the determined inclination angle, but the mode depends on accurately determining the inclination angle, the determined inclination angle is influenced by various factors, the accuracy is difficult to determine, and the correction effect is further influenced.
Along with the development of deep learning technology, various schemes for accurately identifying the inclination angle of an inclined object by using a deep learning training model are layered, but because the deep learning needs to collect and synthesize a large amount of training data and mark the angle of the object in each picture, a convolutional neural network is trained to predict the angle. This solution requires a large labeling cost, labor cost, time cost, and generalization performance cannot be guaranteed.
How to achieve good correction effect and improve correction efficiency is a technical problem to be solved by the technicians in the field.
Disclosure of Invention
The invention aims to provide an object inclination correction method, an object inclination correction device, electronic equipment and a storage medium, which can efficiently achieve a good correction effect.
Embodiments of the invention may be implemented as follows:
in a first aspect, the present invention provides a method of correcting an inclination of an object, the method comprising:
acquiring a plurality of inclined images of a preset object, wherein the inclined angles of the preset object in the plurality of inclined images are different and are all in a preset range, and the difference between any two adjacent inclined angles is a preset difference value;
calculating a projection histogram of each inclined image in a preset direction, wherein the projection histogram of each inclined image comprises a plurality of columns of pixel points of each inclined image in the preset direction, and the value of each column is the number of effective pixel points belonging to the preset object on the column;
taking the serial number of each column of the projection histogram of each inclined image as a moment and taking the value of each column as a signal value of the moment to obtain an inclined signal of each inclined image;
performing signal analysis on the inclination signal of each inclination image to obtain the inclination signal intensity of each inclination signal;
and determining a correction image meeting a preset correction requirement from the plurality of inclined images according to the intensity of each inclined signal.
In an alternative embodiment, the step of performing signal analysis on the tilt signal of each of the tilt images to obtain the tilt signal intensity of each of the tilt signals includes:
performing time domain and frequency domain analysis on each inclined signal to obtain a time domain and frequency domain distribution matrix of each inclined signal, wherein a row of the time domain and frequency domain distribution matrix of each inclined signal represents a time domain, a column of the time domain and frequency domain, and each element in the time domain and frequency domain distribution matrix of each inclined signal represents a signal value of each inclined signal corresponding to the time domain and the frequency domain;
and superposing the signal values of each inclined signal according to the time domain and frequency domain distribution matrix of each inclined signal to obtain the inclined signal intensity of each inclined signal.
In an alternative embodiment, the step of superposing the signal values of each of the oblique signals according to the time domain and frequency domain distribution matrix of each of the oblique signals to obtain the oblique signal strength of each of the oblique signals includes:
for any target tilt signal, superposing signal values of each column of a time domain frequency domain distribution matrix of the target tilt signal to obtain a target row vector, wherein an element value of each column in the target row vector is a sum of signal values of each column in the time domain frequency domain distribution matrix of the target tilt signal;
And taking the maximum element value in the target row vector as the tilt signal intensity of the target tilt signal, and finally obtaining the tilt signal intensity of each tilt signal.
In an alternative embodiment, the step of superposing the signal values of each of the oblique signals according to the time domain and frequency domain distribution matrix of each of the oblique signals to obtain the oblique signal strength of each of the oblique signals includes:
for any target tilt signal, superposing signal values of each column of a time domain frequency domain distribution matrix of the target tilt signal to obtain a target row vector, wherein an element value of each column in the target row vector is a sum of signal values of each column in the time domain frequency domain distribution matrix of the target tilt signal;
and taking the sum of element values of all columns in the target row vector as the tilt signal intensity of the target tilt signal, and finally obtaining the tilt signal intensity of each tilt signal.
In an alternative embodiment, the step of acquiring a plurality of oblique images of the preset object includes:
acquiring an original image of the preset object;
obtaining a plurality of inclination angles within the preset range according to the preset range and the preset difference value;
And performing perspective transformation on the original image according to each inclination angle to obtain an inclination image of a preset object inclined at each inclination angle, and finally obtaining a plurality of inclination images of the preset object.
In an alternative embodiment, the step of calculating a projection histogram of each of the oblique images in a preset direction includes:
for any target inclined image, carrying out gray scale processing on the target inclined image to obtain a gray scale image of the target inclined image;
performing binarization processing on the gray level image of the target inclined image to obtain a binary image of the target inclined image, wherein the value of each pixel point in the binary image of the target inclined image represents whether the pixel point is an effective pixel point belonging to the preset object;
counting the number of effective pixel points of each column of the binary image of the target inclined image in the preset direction to obtain a projection histogram of the target inclined image in the preset direction, and finally obtaining the projection histogram of each inclined image in the preset direction.
In an alternative embodiment, the step of determining a corrected image satisfying a preset correction requirement from the plurality of inclined images according to each of the inclined signal intensities includes:
Taking the inclined image with the maximum inclined signal intensity as a primary selection image;
if the primary selected image meets the preset correction requirement, taking the primary selected image as the correction image;
if the initial image does not meet the preset correction requirement, the preset range and the preset difference value are redetermined according to the preset difference value, the redetermined preset range is smaller than the preset range which is determined last time, and the redetermined preset difference value is not larger than the preset difference value which is determined last time;
redetermining a plurality of inclination angles according to the redetermined preset range and the redetermined preset difference value;
performing perspective transformation on the initially selected image according to each redetermined inclination angle to obtain a plurality of new inclination images;
calculating the new inclination signal intensity of each inclination image, and returning to the step of taking the inclination image with the maximum inclination signal intensity as a primary selection image until the correction image is determined.
In a second aspect, the present invention provides a subject tilt correction device, the device comprising:
the acquisition module is used for acquiring a plurality of inclined images of a preset object, wherein the inclined angles of the preset object in the plurality of inclined images are different and are all in a preset range, and the difference between any two adjacent inclined angles is a preset difference value;
The computing module is used for computing a projection histogram of each inclined image in a preset direction, wherein the projection histogram of each inclined image comprises a plurality of columns of pixel points of each inclined image in the preset direction, and the value of each column is the number of effective pixel points belonging to the preset object on the column;
the analysis module is used for obtaining a tilt signal of each tilt image by taking the serial number of each column of the projection histogram of each tilt image as a moment and taking the value of each column as a signal value of the moment;
the analysis module is further used for carrying out signal analysis on the inclination signal of each inclination image to obtain the inclination signal intensity of each inclination signal;
and the determining module is used for determining a correction image meeting a preset correction requirement from the plurality of inclined images according to the intensity of each inclined signal.
In a third aspect, the present invention provides an electronic device comprising a processor and a memory, the memory being configured to store a program, the processor being configured to implement the object tilt correction method according to any one of the preceding embodiments when the program is executed.
In a fourth aspect, the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the object tilt correction method of any of the preceding embodiments.
Compared with the prior art, the embodiment of the invention converts each inclined image into the inclined signal by calculating the projection histogram of each inclined image in the preset direction, analyzes each inclined signal to obtain the inclined signal intensity of each inclined signal, and determines the corrected image meeting the preset correction requirement according to each inclined signal intensity.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is an exemplary diagram of normal fonts and font tilts caused by different situations provided in an embodiment of the present invention.
Fig. 2 is a flowchart illustrating an object tilt correction method according to an embodiment of the present invention.
Fig. 3 is an exemplary diagram of a vertical projection histogram at different angles according to an embodiment of the present invention.
Fig. 4 is a diagram illustrating vertical projection histograms with different tilt angles and a tilt signal intensity distribution according to an embodiment of the present invention.
Fig. 5 is a block diagram illustrating an object tilt correction device according to an embodiment of the present invention.
Fig. 6 is a block diagram of an electronic device according to an embodiment of the present invention.
Icon: 10-an electronic device; 11-a processor; 12-memory; 13-bus; 100-a subject tilt correction device; 110-an acquisition module; 120-a computing module; 130-an analysis module; 140-determination module.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.
In the description of the present invention, it should be noted that, if the terms "upper", "lower", "inner", "outer", and the like indicate an azimuth or a positional relationship based on the azimuth or the positional relationship shown in the drawings, or the azimuth or the positional relationship in which the inventive product is conventionally put in use, it is merely for convenience of describing the present invention and simplifying the description, and it is not indicated or implied that the apparatus or element referred to must have a specific azimuth, be configured and operated in a specific azimuth, and thus it should not be construed as limiting the present invention.
Furthermore, the terms "first," "second," and the like, if any, are used merely for distinguishing between descriptions and not for indicating or implying a relative importance.
It should be noted that the features of the embodiments of the present invention may be combined with each other without conflict.
Taking the text as an example of an object that needs to be corrected, there are various reasons for generating text tilt for the text: the text itself may be italics, the italics have a fixed inclination angle, or when the camera shoots the text, the shooting angle is inclined, so that the text in the shot image is inclined, or the text is caused by handwriting habit, please refer to fig. 1, fig. 1 is an exemplary diagram of normal fonts and font inclination caused by different conditions provided by the embodiment of the present invention.
For correction of oblique text, one way is by image processing, which is at least two specific implementations: (1) clustering mode: performing operations such as gray level binarization and the like on an image of the text, horizontally blurring, clustering angles of all line segments in the image to obtain an angle value with the highest probability, and performing perspective change on the image through the angle to obtain a normal text; (2) a priori knowledge manner: and obtaining the inclination angle of the text according to the priori knowledge, and carrying out corresponding rotation correction on the image of the text to correct the text so as to align the text line with the horizontal line.
The image processing method depends on accurately determining the inclination angle, and the inclination angle is determined by various factors, so that the accuracy of the determination is difficult to determine, and the correction effect is further affected.
In addition to the image processing mode, with the development of deep learning, a deep learning processing mode also appears, and at least the following specific implementation modes exist: (1) means for tilt angle detection network: and marking the inclined text and the normal text to obtain angle information, and training by designing a proper convolutional neural network to obtain an inclined angle detection model. The angle of the inclined text can be predicted through the model, and correction is carried out according to the prediction result; (2) manner of tilting the text correction network: labeling the inclined text and the normal text to obtain a corrected model, and directly generating corrected text from the inclined text by designing a proper image generation network; (3) means for enhancing the text recognition network: in training an OCR (Optical Character Recognition ) model, samples of skewed text may be added in a data enhancement manner so that the model better adapts to text at different angles.
In the image processing mode, gray scale, binarization, expansion corrosion and statistics of all angles in the image of the text are carried out, and the statistical value is analyzed to calculate the angle value with the maximum possibility. However, the method relies on the binarization effect seriously, when facing to low-quality images, adhesion can be generated between texts, separation of texts and backgrounds can be greatly disturbed, and finally the reliability of the statistical angle value is low.
In the deep learning processing mode, a large amount of training data is collected and synthesized, the angle of the text in each picture is marked, and the training convolutional neural network predicts the angle of the text. This solution requires a large labeling cost, labor cost, time cost, and generalization performance cannot be guaranteed. Factors such as background change, font change, text length and the like all affect the model, and the data set needs to be reproduced and retrained under different scenes.
In view of this, the embodiment of the present invention discards simple statistical analysis, and the inventor develops a new approach, and after careful study and deep analysis on the projection results of the text images at various angles of inclination, the inventor finds that: the more obvious difference between the projection result of the inclined character and the projection result of the normal character is the difference between the wave crest and the wave trough. The height difference of the peaks and the troughs of the normal characters is larger, the troughs are more obvious, and the troughs have more obvious intervals (the intervals are intervals between characters); and the inclined character causes the trough to rise due to inclination, the height difference is reduced, and the trough interval is less obvious. Therefore, the inventors contemplate treating the curve of the projection histogram as a discrete signal. The time domain analysis of the signal can observe the change condition of parameters such as amplitude, frequency, period, phase and the like of the signal along with time, and the frequency domain analysis can provide information of the signal on the frequency domain, so that the frequency components and the frequency characteristics of the signal can be revealed. And calculating the signal intensity of the inclined image by adopting a time domain and frequency domain analysis method in signal processing, and determining the optimal correction image according to the signal intensity. Therefore, statistical errors caused by image quality can be well avoided, huge data requirements are deeply learned, and finally, good correction effect is efficiently achieved.
It should be noted that, the method for correcting object inclination provided in the embodiment of the present invention can be applied not only to angle measurement correction of single characters, but also to angle measurement correction of document images, and also to correction of objects such as stamps and trademarks with characters or characters, which will be described in detail below.
Referring to fig. 2, fig. 2 is a flowchart illustrating a method for correcting an object tilt according to an embodiment of the invention, the method includes the following steps:
step S101, a plurality of inclined images of a preset object are obtained, the inclined angles of the preset object in the plurality of inclined images are different and are all in a preset range, and the difference between any two adjacent inclined angles is a preset difference value.
In this embodiment, the preset object may be a text, or may be a stamp, trademark, etc. with the text. The oblique image may be obtained by performing perspective transformation on an original image of a preset object according to different oblique angles, for example, the preset range is-45 ° to 45 °, the preset difference value is 10 °, and the oblique angle is: -45 °, -35 °, -25 °, -15 °, -5 °, 15 °, 25 °, 35 °, 45 °, an inclination angle being transformed to obtain an inclination image, i.e. the inclination image is 10.
Step S102, calculating a projection histogram of each inclined image in a preset direction, wherein the projection histogram of each inclined image comprises a plurality of columns of pixel points of each inclined image in the preset direction, and the value of each column is the number of effective pixel points belonging to a preset object on the column.
In this embodiment, the preset direction may be a vertical direction or a horizontal direction, and is specifically determined according to the direction of the preset object, for example, when the preset object is a character, and the character direction is a vertical direction, and there is an interval in the horizontal direction, the preset direction is a vertical direction, and when the preset object is a document including a plurality of lines, there is an interval in the vertical direction between the lines, the preset direction is a horizontal direction. Taking the projection histogram in the vertical direction as an example, each column in the projection histogram corresponds to each column of pixels of the oblique image, the value of each column in the projection histogram is the number of pixels belonging to the preset object in a corresponding column of pixels of the oblique image, for example, if the oblique image is 100 columns, the projection histogram in the vertical direction also includes 100 columns, and the value of the 1 st column in the projection histogram is the number of pixels belonging to the preset object in the first column of pixels in the oblique image.
Step S103, the serial number of each column of the projection histogram of each inclined image is taken as the moment, and the value of each column is taken as the signal value of the moment, so as to obtain the inclined signal of each inclined image.
In this embodiment, the oblique image is converted into an oblique signal, so that the oblique signal is analyzed by a signal analysis means, and finally, the analysis of the oblique image is realized.
Step S104, performing signal analysis on the inclination signal of each inclination image to obtain the inclination signal intensity of each inclination signal.
In this embodiment, the signal analysis may be a time domain analysis, a frequency domain analysis, a time domain frequency domain analysis. According to the previous findings of the inventors, the vertical projection histogram of the normal text has a higher peak value, and thus it is known that the tilt signal intensity of the tilt signal of the projection histogram of the normal text is also larger, and therefore, a preset object having a smaller tilt angle in the tilt image can be determined from the tilt signal intensity of the tilt signal.
Step S105, according to each inclined signal intensity, determining a correction image meeting a preset correction requirement from a plurality of inclined images.
In this embodiment, according to the actual scene, the correction image satisfying the preset correction requirement may be directly determined from the multiple oblique images, or a candidate image may be selected from the multiple oblique images, then the inclination angle is redetermined based on the candidate image by using the redetermined preset range and the redetermined preset difference value, and then each new inclination angle is performed on the candidate image to obtain a new oblique image, and then the steps S101-S105 are iterated until the correction image satisfying the preset correction requirement is finally obtained.
In this embodiment, the preset correction requirement may be that the preset range reaches the expected range, or that the number of times of redefining the preset range reaches the preset number of times, etc., and it needs to be noted that the more the number of iterations, the better the correction effect of the obtained correction image, but the longer the processing time is, and the corresponding preset correction requirement may be set according to the needs of the actual scene.
According to the method provided by the embodiment, the projection histogram of the image is converted into the signal, and the correction image meeting the preset correction requirement is determined by utilizing the signal analysis technology, so that a good correction effect is achieved, and the correction efficiency is improved.
In this embodiment, in order to obtain a plurality of oblique images of a preset object, this embodiment provides an implementation manner:
firstly, acquiring an original image of a preset object;
secondly, obtaining a plurality of inclination angles within a preset range according to the preset range and the preset difference value;
finally, perspective transformation is carried out on the original image according to each inclination angle, so that an inclination image of the preset object inclined at each inclination angle is obtained, and finally a plurality of inclination images of the preset object are obtained.
In this embodiment, considering the effect of correction, the maximum preset range is usually set to-45 ° to 45 °, and of course, a range larger or smaller than the maximum preset range may be set specifically according to actual needs, where the smaller the tilt angle setting is, the larger the number of tilt images, the longer the processing time is, but the finer the obtained correction result is, the larger the tilt angle setting is, the smaller the number of tilt images is, the higher the processing efficiency is, and the obtained correction result is lower in relative fineness.
In an alternative embodiment, the manner of calculating the projection histogram in the preset direction is the same for each oblique image, and this embodiment describes a manner of calculating the projection histogram in the preset direction with any target oblique image:
firstly, carrying out gray scale processing on a target inclined image to obtain a gray scale image of the target inclined image;
in this embodiment, the target oblique image may be processed into a single-channel gray-scale image by performing gray-scale processing on the target oblique image.
Secondly, carrying out binarization processing on a gray level image of the target inclined image to obtain a binary image of the target inclined image, wherein the value of each pixel point in the binary image of the target inclined image represents whether the pixel point is an effective pixel point belonging to a preset object;
in this embodiment, the binarization processing may be performed on the gray level image of the target oblique image by using an adaptive binarization processing manner, and as an implementation manner, a cv2.adaptive threshold function of an opencv library (the opencv library is a cross-platform computer vision and machine learning software library) may be used, and the method for calculating the threshold value uses cv2.adapt_thresh_gausianian_c, where the threshold value is a weighted sum of neighborhood values, and the weight is given by a GAUSSIAN window. The calculation parameters in the self-adaptation process can achieve the purpose only by slightly adjusting according to the environments of different images, and default parameters can be adopted for the parameters in most scenes.
In this embodiment, as an implementation manner, when a pixel belongs to a preset object, the pixel value of the pixel is set to 1, otherwise, the pixel value is set to 0, and of course, other values may be set as long as whether the pixel belongs to the preset object can be distinguished.
It should be noted that, the binarization processing in this embodiment does not need to completely separate the text and the background after binarization, as in other image processing schemes, so that when binarizing, the image needs to be subjected to complex processing such as denoising, noise reduction, contrast stretching, and OTSU algorithm determining the binary segmentation threshold of the image, so that an ideal correction effect can be achieved.
Finally, counting the number of effective pixel points of each column of the binary image of the target inclined image in the preset direction to obtain a projection histogram of the target inclined image in the preset direction, and finally obtaining the projection histogram of each inclined image in the preset direction.
In this embodiment, taking the preset direction as the vertical direction as an example, the projection histogram in the vertical direction counts the number of black pixels in the vertical direction of the target oblique image column by column, and the distribution condition of the projection histogram is represented in a plane rectangular coordinate system in the form of the statistics histogram, please refer to fig. 3, fig. 3 is an example diagram of the vertical projection histogram with different angles provided in the embodiment of the present invention, it can be seen from fig. 3 that the text with the minimum oblique angle has the maximum peak-valley height difference and the most obvious valley. The distribution state of black pixels of the projection histogram is regarded as a discrete signal, the picture width of the projection histogram is taken as the length of the discrete signal S, and the value of each column of the projection histogram is taken as the instantaneous value of the signal at the moment, so that the inclination signal corresponding to the target inclination image is obtained.
In an alternative embodiment, one implementation of deriving the tilt signal strength of each tilt signal is:
firstly, carrying out time domain and frequency domain analysis on each inclined signal to obtain a time domain and frequency domain distribution matrix of each inclined signal, wherein a row of the time domain and frequency domain distribution matrix of each inclined signal represents a time domain, a column represents a frequency domain, and each element in the time domain and frequency domain distribution matrix of each inclined signal represents a signal value of each inclined signal corresponding to the time domain and the frequency domain;
In this embodiment, the time domain frequency domain analysis includes, but is not limited to, analysis methods such as fourier Transform, wigner-Ville distribution (i.e. WVD distribution), and in contrast, the conventional fourier Transform analysis method has inherent time-frequency uncertainty, which affects the accuracy of information of signals in time domain and frequency domain, while the WVD distribution can capture transient characteristics of signals in different time and frequency by using autocorrelation functions of time domain and frequency domain, and provide high-resolution time domain frequency domain information, so that the uncertainty existing in the conventional fourier Transform can be overcome, and accurate time domain frequency domain analysis of signals can be implemented.
Referring to fig. 4, fig. 4 is an exemplary diagram of vertical projection histograms and oblique signal intensity distribution with different oblique angles according to an embodiment of the present invention, and as can be seen from fig. 4, the oblique signal intensity of a normal text is stronger and the intensity change is more obvious, so that a corrected image with an oblique angle meeting the requirement can be determined according to the oblique signal intensity.
And secondly, superposing the signal value of each inclined signal according to the time domain and frequency domain distribution matrix of each inclined signal to obtain the inclined signal intensity of each inclined signal.
In an alternative embodiment, at least two manners of obtaining the inclination signal intensity of the inclination signal by superposing the signal values of the inclination signals may be selected according to the needs of the actual scene, or the corresponding inclination signal intensities may be obtained in two manners, and then final judgment may be made according to the two inclination signal intensities obtained in two manners, for example, the largest inclination signal intensity is selected as the final inclination signal intensity, or the average value of the two inclination signal intensities is used as the final inclination signal intensity.
Once the superposition mode is selected, the superposition mode of each tilt signal is performed according to the selected superposition mode, and for convenience of description, two superposition modes are respectively introduced below with any target tilt signal:
mode one:
firstly, superposing signal values of each column of a time domain frequency domain distribution matrix of a target inclined signal to obtain a target row vector, wherein the element value of each column in the target row vector is the sum of the signal values of each column in the time domain frequency domain distribution matrix of the target inclined signal;
and secondly, taking the maximum element value in the target row vector as the tilt signal intensity of the target tilt signal, and finally obtaining the tilt signal intensity of each tilt signal.
For example, the time domain frequency domain distribution matrix of the target tilt signal a is:
and superposing the signal values of each column of the matrix to obtain a target row vector (6,8,7,11), wherein the maximum element value in the target row vector is 11, and the inclined signal intensity of a is 11.
Mode two:
firstly, superposing signal values of each column of a time domain frequency domain distribution matrix of a target inclined signal to obtain a target row vector, wherein the element value of each column in the target row vector is the sum of the signal values of each column in the time domain frequency domain distribution matrix of the target inclined signal;
and secondly, taking the sum of element values of all columns in the target row vector as the tilt signal intensity of the target tilt signal, and finally obtaining the tilt signal intensity of each tilt signal.
For example, if the target row vector of the target tilt signal a is (6,8,7,11), the element values of all columns in the target row vector are summed to obtain 32, and if the tilt signal strength of a is 32.
In an alternative embodiment, according to each tilt signal intensity, the implementation manner of determining the corrected image meeting the preset correction requirement from the plurality of tilt images is as follows:
firstly, taking an inclined image with the maximum inclined signal intensity as a primary selection image;
In this embodiment, according to the needs of the actual scene, if the efficiency is the scene with priority, the primary selected image may be directly used as the correction image, if the correction effect is the scene with priority, the processing with finer granularity may be performed again based on the primary selected image, so as to achieve the better correction effect, under the normal scene, the primary screening processing may be performed by setting the preset range with coarser granularity and the preset difference value, so as to obtain the primary selected image, and then the fine screening processing may be performed by gradually setting the preset range with finer granularity and the preset difference value based on the primary selected image, so as to obtain the correction image with better correction effect under the condition of ideal efficiency. For example, the first time: the preset range is set to be-45 degrees to 45 degrees, the preset difference value is 10 degrees, and the second time: the preset range is set to be-10 degrees to 10 degrees, and the preset difference value is 2 degrees.
Secondly, if the primary selected image meets the preset correction requirement, taking the primary selected image as a correction image;
thirdly, if the initially selected image does not meet the preset correction requirement, the preset range and the preset difference value are redetermined according to the preset difference value, the redetermined preset range is smaller than the preset range which is determined last time, and the redetermined preset difference value is not larger than the preset difference value which is determined last time;
In this embodiment, as an implementation manner, the preset range may be determined according to a preset difference, for example, the preset difference is 10 °, the redetermined preset range is-10 ° to 10 °, and in order to make the screening granularity smaller, the redetermined preset difference is set to be not greater than the preset difference determined last time, so that the purpose of selecting finer granularity can be achieved once each iteration, and the effect of the finally obtained corrected image is best.
Fourth, redetermining a plurality of inclination angles according to the redetermined preset range and the redetermined preset difference value;
fifthly, performing perspective transformation on the initially selected image according to each redetermined inclination angle to obtain a new plurality of inclination images;
sixth, calculate the new oblique signal intensity of each oblique image, and return to the step of taking the oblique image with the largest oblique signal intensity as the primary image, until confirm and correct the picture.
In this embodiment, the step of calculating the new tilt signal intensity of each tilt image may refer to the processing of step S101 to step S104 and the description of the specific implementation of each step described in this embodiment, and one way of describing the iterative processing may be:
Step1: acquiring a plurality of inclined images of a preset object, wherein the inclined angles of the preset object in the plurality of inclined images are different and are all in a preset range, and the difference between any two adjacent inclined angles is a preset difference value;
step2: calculating a projection histogram of each inclined image in a preset direction, wherein the projection histogram of each inclined image comprises a plurality of columns of pixel points of each inclined image in the preset direction, and the value of each column is the number of effective pixel points belonging to a preset object on the column;
step3: taking the serial number of each column of the projection histogram of each inclined image as the moment and taking the value of each column as the signal value of the moment to obtain an inclined signal of each inclined image;
step4: performing signal analysis on the inclination signal of each inclination image to obtain the inclination signal intensity of each inclination signal;
step5: taking the inclined image with the maximum inclined signal intensity as a primary selection image;
step6: if the initially selected image does not meet the preset correction requirement, the preset range and the preset difference value are redetermined according to the preset difference value, the redetermined preset range is smaller than the preset range which is determined last time, and the redetermined preset difference value is not larger than the preset difference value which is determined last time;
Step7: redetermining a plurality of inclination angles according to the redetermined preset range and the redetermined preset difference value;
step8: and performing perspective transformation on the primary selected image according to each redetermined inclination angle to obtain a plurality of new inclination images, and returning to Step2 until the corrected image is determined.
In order to perform the respective steps of the above examples and various possible embodiments, an implementation of the object inclination correction device is given below. Referring to fig. 5, fig. 5 is a block diagram illustrating an object tilt correction device according to an embodiment of the invention. It should be noted that, the basic principle and the technical effects of the object inclination correction device 100 provided in this embodiment are the same as those of the above embodiment, and for brevity, the description of this embodiment is not mentioned.
The subject tilt correction device includes an acquisition module 110, a calculation module 120, an analysis module 130, and a determination module 140.
The acquisition module 110 is configured to acquire a plurality of inclination images of a preset object, wherein the inclination angles of the preset object in the plurality of inclination images are different and are all within a preset range, and a difference between any two adjacent inclination angles is a preset difference value;
the calculating module 120 is configured to calculate a projection histogram of each oblique image in a preset direction, where the projection histogram of each oblique image includes a plurality of columns of pixels of each oblique image in the preset direction, and a value of each column is a number of valid pixels belonging to a preset object on the column;
An analysis module 130, configured to obtain a tilt signal of each tilt image with a sequence number of each column of the projection histogram of each tilt image as a time and a value of each column as a signal value of the time;
the analysis module 130 is further configured to perform signal analysis on the tilt signal of each tilt image, so as to obtain a tilt signal intensity of each tilt signal;
the determining module 140 is configured to determine, according to each tilt signal intensity, a corrected image that meets a preset correction requirement from the plurality of tilt images.
In an alternative embodiment, the analysis module 130 is specifically configured to: performing time domain and frequency domain analysis on each inclined signal to obtain a time domain and frequency domain distribution matrix of each inclined signal, wherein a row of the time domain and frequency domain distribution matrix of each inclined signal represents a time domain, a column represents a frequency domain, and each element in the time domain and frequency domain distribution matrix of each inclined signal represents a signal value of each inclined signal corresponding to the time domain and the frequency domain; and superposing the signal value of each inclined signal according to the time domain frequency domain distribution matrix of each inclined signal to obtain the inclined signal intensity of each inclined signal.
In an alternative embodiment, the analysis module 130 is specifically configured to, when stacking the signal values of each tilt signal according to the time domain and frequency domain distribution matrix of each tilt signal to obtain the tilt signal strength of each tilt signal: for any target tilt signal, superposing the signal values of each column of the time domain frequency domain distribution matrix of the target tilt signal to obtain a target row vector, wherein the element value of each column in the target row vector is the sum of the signal values of each column in the time domain frequency domain distribution matrix of the target tilt signal; and taking the maximum element value in the target row vector as the tilt signal intensity of the target tilt signal, and finally obtaining the tilt signal intensity of each tilt signal.
In an alternative embodiment, the analysis module 130 is specifically configured to, when stacking the signal values of each tilt signal according to the time domain and frequency domain distribution matrix of each tilt signal to obtain the tilt signal strength of each tilt signal: for any target tilt signal, superposing the signal values of each column of the time domain frequency domain distribution matrix of the target tilt signal to obtain a target row vector, wherein the element value of each column in the target row vector is the sum of the signal values of each column in the time domain frequency domain distribution matrix of the target tilt signal; and taking the sum of element values of all columns in the target row vector as the tilt signal intensity of the target tilt signal, and finally obtaining the tilt signal intensity of each tilt signal.
In an alternative embodiment, the obtaining module 110 is specifically configured to: acquiring an original image of a preset object; obtaining a plurality of inclination angles within a preset range according to the preset range and the preset difference value; and performing perspective transformation on the original image according to each inclination angle to obtain an inclination image of the preset object inclined at each inclination angle, and finally obtaining a plurality of inclination images of the preset object.
In an alternative embodiment, the computing module 120 is specifically configured to: for any target inclined image, carrying out gray processing on the target inclined image to obtain a gray image of the target inclined image; performing binarization processing on a gray level image of a target inclined image to obtain a binary image of the target inclined image, wherein a value of each pixel point in the binary image of the target inclined image represents whether the pixel point is an effective pixel point belonging to a preset object; counting the number of effective pixel points of each column of the binary image of the target inclined image in the preset direction to obtain a projection histogram of the target inclined image in the preset direction, and finally obtaining the projection histogram of each inclined image in the preset direction.
In an alternative embodiment, the calculating module 120 is specifically configured to, when determining, according to each tilt signal intensity, a corrected image that meets a preset correction requirement from the plurality of tilt images: taking the inclined image with the maximum inclined signal intensity as a primary selection image; if the primary selected image meets the preset correction requirement, taking the primary selected image as a correction image; if the initially selected image does not meet the preset correction requirement, the preset range and the preset difference value are redetermined according to the preset difference value, the redetermined preset range is smaller than the preset range which is determined last time, and the redetermined preset difference value is not larger than the preset difference value which is determined last time; redetermining a plurality of inclination angles according to the redetermined preset range and the redetermined preset difference value; performing perspective transformation on the primary selected image according to each redetermined inclination angle to obtain a plurality of new inclination images; calculating the new inclination signal intensity of each inclination image, and returning to the step of taking the inclination image with the maximum inclination signal intensity as the primary selection image until the correction image is determined.
Referring to fig. 6, fig. 6 is a schematic block diagram of the electronic device 10 according to the embodiment of the present invention, where the electronic device 10 includes a processor 11, a memory 12, and a bus 13. The processor 11 and the memory 12 communicate via a bus 13.
The processor 11 may be an integrated circuit chip with signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in the processor 11 or by instructions in the form of software. The processor 11 may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU for short), a network processor (Network Processor, NP for short), and the like; but may also be a Digital Signal Processor (DSP), application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components.
The memory 12 is used for storing a program, for example, an object tilt correction device in the embodiment of the present invention, and the object tilt correction device includes at least one software function module that may be stored in the memory 12 in the form of software or firmware (firmware), and the processor 11 executes the program after receiving the execution instruction to implement the object tilt correction method in the embodiment of the present invention.
The memory 12 may include high-speed random access memory (RAM: random Access Memory) and may also include non-volatile memory (nonvolatile memory). Alternatively, the memory 12 may be a storage device built into the processor 11, or may be a storage device independent of the processor 11.
The bus 13 may be an ISA bus, a PCI bus, an EISA bus, or the like. Fig. 6 is represented by only one double-headed arrow, but does not represent only one bus or one type of bus.
An embodiment of the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the object tilt correction method of any of the foregoing embodiments.
In summary, the embodiments of the present invention provide a method, an apparatus, an electronic device, and a storage medium for correcting object inclination, where the method includes: acquiring a plurality of inclined images of a preset object, wherein the inclined angles of the preset object in the plurality of inclined images are different and are all in a preset range, and the difference between any two adjacent inclined angles is a preset difference value; calculating a projection histogram of each inclined image in a preset direction, wherein the projection histogram of each inclined image comprises a plurality of columns of pixel points of each inclined image in the preset direction, and the value of each column is the number of effective pixel points belonging to a preset object on the column; taking the serial number of each column of the projection histogram of each inclined image as the moment and taking the value of each column as the signal value of the moment to obtain an inclined signal of each inclined image; performing signal analysis on the inclination signal of each inclination image to obtain the inclination signal intensity of each inclination signal; and determining a correction image meeting preset correction requirements from the plurality of inclined images according to the intensity of each inclined signal. Compared with the prior art, the embodiment of the invention has at least the following advantages: (1) The projection histogram of the image is converted into a signal, and the correction image meeting the preset correction requirement is determined by utilizing the signal analysis technology, so that a good correction effect is achieved, and the correction efficiency is improved; (2) The inclination signal of the inclination image is subjected to signal analysis, so that the intensity of the signal changing along with the inclination angle can be more accurately and stably reflected, the judgment result is more accurate, and the finally determined correction image is more accurate; (3) And the method comprises the steps of firstly setting a preset range and a preset difference value of coarser granularity by using a multi-iteration mode, performing primary screening treatment to obtain a primary selected image, then gradually setting the preset range and the preset difference value of finer granularity based on the primary selected image, performing fine screening treatment, and finally obtaining a corrected image with better correction effect under the condition of ideal efficiency.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any changes or substitutions easily contemplated by those skilled in the art within the scope of the present invention should be included in the present invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (8)
1. A method of correcting tilt of an object, the method comprising:
acquiring a plurality of inclined images of a preset object, wherein the inclined angles of the preset object in the plurality of inclined images are different and are all in a preset range, and the difference between any two adjacent inclined angles is a preset difference value;
calculating a projection histogram of each inclined image in a preset direction, wherein the projection histogram of each inclined image comprises a plurality of columns of pixel points of each inclined image in the preset direction, and the value of each column is the number of effective pixel points belonging to the preset object on the column;
taking the serial number of each column of the projection histogram of each inclined image as a moment and taking the value of each column as a signal value of the moment to obtain an inclined signal of each inclined image;
Performing signal analysis on the inclination signal of each inclination image to obtain the inclination signal intensity of each inclination signal;
according to the intensity of each inclined signal, determining a correction image meeting a preset correction requirement from the plurality of inclined images;
the step of analyzing the tilt signal of each tilt image to obtain the tilt signal intensity of each tilt signal includes:
performing time domain and frequency domain analysis on each inclined signal to obtain a time domain and frequency domain distribution matrix of each inclined signal, wherein a row of the time domain and frequency domain distribution matrix of each inclined signal represents a time domain, a column of the time domain and frequency domain, and each element in the time domain and frequency domain distribution matrix of each inclined signal represents a signal value of each inclined signal corresponding to the time domain and the frequency domain;
superposing signal values of each inclined signal according to a time domain and frequency domain distribution matrix of each inclined signal to obtain inclined signal intensity of each inclined signal;
the step of determining a corrected image satisfying a preset correction requirement from the plurality of inclined images according to each inclined signal intensity includes:
taking the inclined image with the maximum inclined signal intensity as a primary selection image;
If the primary selected image meets the preset correction requirement, taking the primary selected image as the correction image;
if the initial image does not meet the preset correction requirement, the preset range and the preset difference value are redetermined according to the preset difference value, the redetermined preset range is smaller than the preset range which is determined last time, and the redetermined preset difference value is not larger than the preset difference value which is determined last time;
redetermining a plurality of inclination angles according to the redetermined preset range and the redetermined preset difference value;
performing perspective transformation on the initially selected image according to each redetermined inclination angle to obtain a plurality of new inclination images;
calculating the new inclination signal intensity of each inclination image, and returning to the step of taking the inclination image with the maximum inclination signal intensity as a primary selection image until the correction image is determined.
2. The method of claim 1, wherein the step of superimposing the signal values of each of the tilt signals according to the time-domain frequency-domain distribution matrix of each of the tilt signals to obtain the tilt signal intensity of each of the tilt signals comprises:
for any target tilt signal, superposing signal values of each column of a time domain frequency domain distribution matrix of the target tilt signal to obtain a target row vector, wherein an element value of each column in the target row vector is a sum of signal values of each column in the time domain frequency domain distribution matrix of the target tilt signal;
And taking the maximum element value in the target row vector as the tilt signal intensity of the target tilt signal, and finally obtaining the tilt signal intensity of each tilt signal.
3. The method of claim 1, wherein the step of superimposing the signal values of each of the tilt signals according to the time-domain frequency-domain distribution matrix of each of the tilt signals to obtain the tilt signal intensity of each of the tilt signals comprises:
for any target tilt signal, superposing signal values of each column of a time domain frequency domain distribution matrix of the target tilt signal to obtain a target row vector, wherein an element value of each column in the target row vector is a sum of signal values of each column in the time domain frequency domain distribution matrix of the target tilt signal;
and taking the sum of element values of all columns in the target row vector as the tilt signal intensity of the target tilt signal, and finally obtaining the tilt signal intensity of each tilt signal.
4. The object tilt correction method of claim 1, wherein the step of acquiring a plurality of tilt images of the predetermined object comprises:
acquiring an original image of the preset object;
Obtaining a plurality of inclination angles within the preset range according to the preset range and the preset difference value;
and performing perspective transformation on the original image according to each inclination angle to obtain an inclination image of a preset object inclined at each inclination angle, and finally obtaining a plurality of inclination images of the preset object.
5. The object tilt correction method of claim 1, wherein the step of calculating a projection histogram of each of the tilt images in a predetermined direction comprises:
for any target inclined image, carrying out gray scale processing on the target inclined image to obtain a gray scale image of the target inclined image;
performing binarization processing on the gray level image of the target inclined image to obtain a binary image of the target inclined image, wherein the value of each pixel point in the binary image of the target inclined image represents whether the pixel point is an effective pixel point belonging to the preset object;
counting the number of effective pixel points of each column of the binary image of the target inclined image in the preset direction to obtain a projection histogram of the target inclined image in the preset direction, and finally obtaining the projection histogram of each inclined image in the preset direction.
6. A subject tilt correction device, the device comprising:
the acquisition module is used for acquiring a plurality of inclined images of a preset object, wherein the inclined angles of the preset object in the plurality of inclined images are different and are all in a preset range, and the difference between any two adjacent inclined angles is a preset difference value;
the computing module is used for computing a projection histogram of each inclined image in a preset direction, wherein the projection histogram of each inclined image comprises a plurality of columns of pixel points of each inclined image in the preset direction, and the value of each column is the number of effective pixel points belonging to the preset object on the column;
the analysis module is used for obtaining a tilt signal of each tilt image by taking the serial number of each column of the projection histogram of each tilt image as a moment and taking the value of each column as a signal value of the moment;
the analysis module is further used for carrying out signal analysis on the inclination signal of each inclination image to obtain the inclination signal intensity of each inclination signal;
the determining module is used for determining correction images meeting preset correction requirements from the plurality of inclined images according to the intensity of each inclined signal;
The analysis module is specifically used for: performing time domain and frequency domain analysis on each inclined signal to obtain a time domain and frequency domain distribution matrix of each inclined signal, wherein a row of the time domain and frequency domain distribution matrix of each inclined signal represents a time domain, a column of the time domain and frequency domain, and each element in the time domain and frequency domain distribution matrix of each inclined signal represents a signal value of each inclined signal corresponding to the time domain and the frequency domain; superposing signal values of each inclined signal according to a time domain and frequency domain distribution matrix of each inclined signal to obtain inclined signal intensity of each inclined signal;
the determining module is specifically configured to: taking the inclined image with the maximum inclined signal intensity as a primary selection image; if the primary selected image meets the preset correction requirement, taking the primary selected image as the correction image; if the initial image does not meet the preset correction requirement, the preset range and the preset difference value are redetermined according to the preset difference value, the redetermined preset range is smaller than the preset range which is determined last time, and the redetermined preset difference value is not larger than the preset difference value which is determined last time; redetermining a plurality of inclination angles according to the redetermined preset range and the redetermined preset difference value; performing perspective transformation on the initially selected image according to each redetermined inclination angle to obtain a plurality of new inclination images; calculating the new inclination signal intensity of each inclination image, and returning to the step of taking the inclination image with the maximum inclination signal intensity as a primary selection image until the correction image is determined.
7. An electronic device comprising a processor and a memory, the memory for storing a program, the processor for implementing the object tilt correction method of any of claims 1-5 when the program is executed.
8. A computer-readable storage medium, characterized in that a computer program is stored thereon, which, when being executed by a processor, implements the object tilt correction method according to any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311330100.6A CN117078913B (en) | 2023-10-16 | 2023-10-16 | Object inclination correction method, device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311330100.6A CN117078913B (en) | 2023-10-16 | 2023-10-16 | Object inclination correction method, device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117078913A CN117078913A (en) | 2023-11-17 |
| CN117078913B true CN117078913B (en) | 2024-02-02 |
Family
ID=88717456
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311330100.6A Active CN117078913B (en) | 2023-10-16 | 2023-10-16 | Object inclination correction method, device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117078913B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003223636A (en) * | 2002-01-30 | 2003-08-08 | Fuji Photo Film Co Ltd | Image processor |
| JP2011090437A (en) * | 2009-10-21 | 2011-05-06 | Sony Corp | Image processor, method, and program |
| WO2011155161A1 (en) * | 2010-06-11 | 2011-12-15 | パナソニック株式会社 | Image processing apparatus and image processing method |
| CN107220644A (en) * | 2017-04-18 | 2017-09-29 | 天津大学 | A kind of ecg scanning image gradient bearing calibration |
| CN108009538A (en) * | 2017-12-22 | 2018-05-08 | 大连运明自动化技术有限公司 | A kind of automobile engine cylinder-body sequence number intelligent identification Method |
| KR20180087532A (en) * | 2017-01-24 | 2018-08-02 | 울산대학교 산학협력단 | An acquisition system of distance information in direction signs for vehicle location information and method |
| CN108648151A (en) * | 2018-05-09 | 2018-10-12 | 电子科技大学 | A kind of license plate horizontally tilted automatic correction method under bad weather |
| CN109409356A (en) * | 2018-08-23 | 2019-03-01 | 浙江理工大学 | A kind of multi-direction Chinese print hand writing detection method based on SWT |
| CN111783773A (en) * | 2020-06-15 | 2020-10-16 | 北京工业大学 | Correction method for angle-oriented inclined wire pole signboard |
| CN113128495A (en) * | 2019-12-30 | 2021-07-16 | 广州慧睿思通科技股份有限公司 | Text inclination correction method and device, storage medium and computer equipment |
| CN113177899A (en) * | 2021-05-25 | 2021-07-27 | 上海海事大学 | Method for correcting text tilt of medical photocopy, electronic device and readable storage medium |
| CN115578695A (en) * | 2022-11-21 | 2023-01-06 | 昆明理工大学 | Water gauge water level machine vision detection method and device with free shooting visual angle |
-
2023
- 2023-10-16 CN CN202311330100.6A patent/CN117078913B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003223636A (en) * | 2002-01-30 | 2003-08-08 | Fuji Photo Film Co Ltd | Image processor |
| JP2011090437A (en) * | 2009-10-21 | 2011-05-06 | Sony Corp | Image processor, method, and program |
| WO2011155161A1 (en) * | 2010-06-11 | 2011-12-15 | パナソニック株式会社 | Image processing apparatus and image processing method |
| KR20180087532A (en) * | 2017-01-24 | 2018-08-02 | 울산대학교 산학협력단 | An acquisition system of distance information in direction signs for vehicle location information and method |
| CN107220644A (en) * | 2017-04-18 | 2017-09-29 | 天津大学 | A kind of ecg scanning image gradient bearing calibration |
| CN108009538A (en) * | 2017-12-22 | 2018-05-08 | 大连运明自动化技术有限公司 | A kind of automobile engine cylinder-body sequence number intelligent identification Method |
| CN108648151A (en) * | 2018-05-09 | 2018-10-12 | 电子科技大学 | A kind of license plate horizontally tilted automatic correction method under bad weather |
| CN109409356A (en) * | 2018-08-23 | 2019-03-01 | 浙江理工大学 | A kind of multi-direction Chinese print hand writing detection method based on SWT |
| CN113128495A (en) * | 2019-12-30 | 2021-07-16 | 广州慧睿思通科技股份有限公司 | Text inclination correction method and device, storage medium and computer equipment |
| CN111783773A (en) * | 2020-06-15 | 2020-10-16 | 北京工业大学 | Correction method for angle-oriented inclined wire pole signboard |
| CN113177899A (en) * | 2021-05-25 | 2021-07-27 | 上海海事大学 | Method for correcting text tilt of medical photocopy, electronic device and readable storage medium |
| CN115578695A (en) * | 2022-11-21 | 2023-01-06 | 昆明理工大学 | Water gauge water level machine vision detection method and device with free shooting visual angle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117078913A (en) | 2023-11-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110414507B (en) | License plate recognition method and device, computer equipment and storage medium | |
| CN110046529B (en) | Two-dimensional code identification method, device and equipment | |
| CN108985137B (en) | License plate recognition method, device and system | |
| CN106156766B (en) | Method and device for generating text line classifier | |
| CN110020692B (en) | Handwriting separation and positioning method based on print template | |
| CN110619333B (en) | Text line segmentation method, text line segmentation device and electronic equipment | |
| CN108108734B (en) | License plate recognition method and device | |
| CN109034158B (en) | License plate recognition method and device and computer equipment | |
| CN109447080B (en) | Character recognition method and device | |
| US10922535B2 (en) | Method and device for identifying wrist, method for identifying gesture, electronic equipment and computer-readable storage medium | |
| CN112861870B (en) | Pointer instrument image correction method, system and storage medium | |
| CN114387591A (en) | License plate recognition method, system, equipment and storage medium | |
| CN113269706B (en) | Laser radar image quality evaluation method, device, equipment and storage medium | |
| CN110942473A (en) | Moving target tracking detection method based on characteristic point gridding matching | |
| CN114078213A (en) | Farmland contour detection method and device based on generation of confrontation network | |
| CN108960247B (en) | Image significance detection method and device and electronic equipment | |
| CN117496560B (en) | Fingerprint line identification method and device based on multidimensional vector | |
| JP6341059B2 (en) | Character recognition device, character recognition method, and program | |
| US10176400B1 (en) | Method and apparatus for locating dot text in an image | |
| CN117078913B (en) | Object inclination correction method, device, electronic equipment and storage medium | |
| US10210414B2 (en) | Object detection system and computer program product | |
| CN112560856A (en) | License plate detection and identification method, device, equipment and storage medium | |
| CN111898408A (en) | Rapid face recognition method and device | |
| CN117152760A (en) | Image processing method, terminal device and storage medium | |
| US20150279039A1 (en) | Object detecting apparatus and method |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240429 Address after: 710000, 1st to 5th floors, Building 15, Qianhai Garden, Silk Road (Xi'an), No. 900 Xitai Road, High tech Zone, Xi'an City, Shaanxi Province Patentee after: XI'AN GLASSSIX NETWORK TECHNOLOGY CO.,LTD. Country or region after: China Address before: No.3, 26 / F, building 3, No.88, Jitai fifth road, hi tech Zone, Chengdu, Sichuan 610000 Patentee before: GLASSSIC TECHNOLOGY (CHENGDU) Co.,Ltd. Country or region before: China |
|
| TR01 | Transfer of patent right |