CN111753830A - A job image correction method and computing device - Google Patents

Abstract

The invention discloses a job image correction method which is suitable for being executed in computing equipment and comprises the following steps: acquiring a job image to be processed, and identifying each character connected domain in the image; dividing one or more character communication domains with similar left-most horizontal coordinates into the same group; determining the operation image to be in single-column layout or multi-column layout and a corresponding column area according to the divided one or more groups; and for each column area, determining one or more standard communication fields with the width reaching a preset condition, and carrying out image correction on the column area based on the text lines of the standard communication fields. The invention also discloses a computing device for executing the method.

Description

A job image correction method and computing device

technical field

The invention relates to the field of image processing, in particular to a job image correction method and a computing device.

Background technique

Homework pictures have certain particularities, generally divided into single-column pictures, double-column pictures, or even three-column or four-column pictures, each column area has a topic. In addition, there will be special cases such as single column as the main part and double column in the job picture. Moreover, the job picture is generally on the workbook, and the user obtains the job picture through this page of work paper, and the workbook may have a certain thickness, resulting in a certain image distortion in the captured image, for example, part of the content on the page is distorted downward, and the lower part of the content is distorted. Upward twist etc. There is another part of the homework picture with less text area, such as geometric figures, cartoon characters and answer areas in the question. Therefore, it is necessary to provide a method for analyzing and correcting the work pictures and for identifying the plate type, so as to confirm each problem area.

SUMMARY OF THE INVENTION

In view of the above problems, the present invention proposes a job image correction method and computing device, in an attempt to solve or at least solve the above problems.

According to one aspect of the present invention, there is provided a job image correction method suitable for execution in a computing device, the method comprising the steps of: acquiring a job image to be processed, identifying each character connected domain in the image; One or more text communication domains with similar horizontal coordinates are divided into the same group; according to the divided one or more groups, it is determined that the job image is a single-column layout or a multi-column layout, and the corresponding column area; and for each column area, One or more standard communication fields whose widths meet a predetermined condition are determined therefrom, and image correction is performed on the column area based on the text lines of the standard communication fields.

Optionally, in the job image correction method according to the present invention, it also includes the step: for a multi-column layout, according to the image correction results of two adjacent column areas, image correction is performed on the transition area between the adjacent column areas. .

Optionally, in the job image correction method according to the present invention, the standard communication domains are the uppermost and the lowermost two character communication domains among the character communication domains whose widths meet a predetermined condition.

Optionally, in the job image correction method according to the present invention, the step of performing image correction on the corresponding column area based on the text line of the standard connectivity domain includes: calculating respectively according to the text line and the image horizontal line of the two standard connectivity domains. Corresponding transformation formulas are obtained, and the column area is corrected according to the two transformation formulas obtained by calculation.

Optionally, in the job image correction method according to the present invention, the step of determining that the job image is a single-column layout or a multi-column layout and a corresponding column area according to one or more divided groups includes: If the number is 1, it is determined that the job image is a single-column layout, and the corresponding column area is the area between the minimum abscissa and the maximum abscissa of all the text connection fields in the job image; if the number of groups is not 1, each grouped rectangular areas, and perform deduplication processing on the plurality of rectangular areas, and the deduplicated rectangular area is the column area.

Optionally, in the job image correction method according to the present invention, the left and right boundaries of the rectangular area are the leftmost and rightmost boundaries of all text communication domains in the corresponding group, and the upper and lower boundaries are the upper and lower boundaries of the job image.

Optionally, in the job image correction method according to the present invention, the step of performing deduplication processing on the plurality of rectangular areas includes: deleting rectangular areas that overlap with more than one rectangular area; Rectangular regions that are added to other rectangular regions, and the rectangular region with the smaller width is removed from the two intersecting rectangular regions.

Optionally, in the job image correction method according to the present invention, before recognizing each text connected domain in the image, the method further includes the steps of: converting the job image into a binary image, and performing morphological expansion processing on the binary image, Make adjacent text on the same line connectable.

Optionally, in the job image correction method according to the present invention, the text communication field is the text communication field of printed characters.

Optionally, in the job image correction method according to the present invention, the leftmost abscissas being close means that the difference between the leftmost abscissas of the text communication domain is less than the first percentage of the page width.

Optionally, in the job image correction method according to the present invention, the method further includes the steps of: detecting a vertical straight line in the job image, and performing vertical correction on the job image according to the included angle between the vertical straight line and the image vertical line; and /or detect a horizontal straight line in the job image, and perform horizontal correction on the job image according to the included angle between the horizontal straight line and the horizontal line of the image.

Optionally, in the job image correction method according to the present invention, the vertical straight line is a straight line whose included angle with the vertical line of the image is less than a predetermined angle and whose height is greater than the second percentage of the image height, or the job image. The vertical border line of the image; the horizontal straight line is the straight line whose angle with the horizontal line of the image is less than the predetermined angle and the straight line height is greater than the second percentage of the image height, or the horizontal border line of the job image;

Optionally, in the job image correction method according to the present invention, the width reaching a predetermined condition means that the width is greater than or equal to 75% of the column width, the first percentage is 8%, the second percentage is 50%, and the predetermined angle is is 15 degrees.

Optionally, in the job image correction method according to the present invention, it also includes the step of dividing each topic area in each column area: for each column area, determine the first text connection domain of each line of text in the area. ; From top to bottom, determine the title line based on the first several character formats, contents and abscissa positions of the determined respective text connection fields in turn; and determine the area of each title according to the ordinate positions of the adjacent two title lines. , and divide each topic area for storage.

Optionally, in the homework image correction method according to the present invention, the determination rule of the question number line includes: the question number is Chinese numerals or Arabic numerals; The punctuation after the title number is the same.

Optionally, in the work image correction method according to the present invention, the method further includes the step of: marking the row where the picture is located as the topic area of the closest question number row above the picture.

According to yet another aspect of the present invention, there is provided a computing device comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by one or more Executed by a plurality of processors, the one or more programs, when executed by the processors, implement the steps of the job image correction method as described above.

According to yet another aspect of the present invention, there is provided a computer-readable storage medium storing one or more programs including instructions that, when executed by a computing device, implement a job image as described above The steps of the calibration method.

According to the technical solution of the present invention, one or more groups are determined according to the leftmost coordinate of each text connected domain in the job image, and by analyzing these groupings, it is determined whether the job image is a single-column image or a multi-column image. The column area of a single-column image is the overall operation area itself, and the column area of a multi-column image is the area of each sub-column. After that, a standard connected domain whose width reaches the standard is selected from each column area, and the image is corrected based on the text line and image horizontal line of the standard connected domain. The invention can not only accurately identify the layout layout of complex homework images, but also correct the distorted homework images, which facilitates subsequent process of subject content identification and subject division.

Moreover, when performing image correction, the present invention selects the two text connected domains with the largest and smallest ordinates as standard connectivity from the text connections whose width reaches the standard (for example, the width of the connected domain is greater than or equal to 75% of the column width) in each column area. area. The transformation formula is calculated according to the text line and horizontal line of the two connected domains. The upper picture is transformed downward and corrected layer by layer according to the upper transformation formula, and the lower picture is transformed upward and corrected layer by layer according to the lower transformation formula. The default middle row has the least distortion. The invention adopts a layer-by-layer correction method from both ends (top and bottom) to the middle, so as to improve the accuracy of image correction and make the corrected picture closer to the flat printing format.

The above description is only an overview of the technical solutions of the present invention, in order to be able to understand the technical means of the present invention more clearly, it can be implemented according to the content of the description, and in order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand , the following specific embodiments of the present invention are given.

Description of drawings

To achieve the above and related objects, certain illustrative aspects are described herein in conjunction with the following description and drawings, which are indicative of the various ways in which the principles disclosed herein may be practiced, and all aspects and their equivalents are intended to be within the scope of the claimed subject matter. The above and other objects, features and advantages of the present disclosure will become more apparent by reading the following detailed description in conjunction with the accompanying drawings. Throughout this disclosure, the same reference numbers generally refer to the same parts or elements.

FIG. 1 shows a structural diagram of a computing device 100 according to an embodiment of the present invention;

FIG. 2 shows a flowchart of a job image correction method 200 according to an embodiment of the present invention;

FIG. 3 shows a schematic diagram of multiple text connected domains in a job image according to another embodiment of the present invention;

FIG. 4 shows a schematic diagram of a plurality of grouped rectangles in a job image according to an embodiment of the present invention;

FIG. 5 and FIG. 6 respectively show schematic diagrams of the mutual relationship between grouped rectangles in a job image according to an embodiment of the present invention;

7 shows a schematic diagram of a column area finally delineated by a job image according to an embodiment of the present invention;

FIG. 8 shows a flow chart of a method 800 for dividing an assignment topic according to an embodiment of the present invention; and

FIG. 9 shows a flow chart of a method for dividing an assignment topic according to another embodiment of the present invention.

Detailed ways

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure will be more thoroughly understood, and will fully convey the scope of the present disclosure to those skilled in the art.

1 is a block diagram of a computing device 100 according to one embodiment of the present invention. In a basic configuration 102 , computing device 100 typically includes system memory 106 and one or more processors 104 . The memory bus 108 may be used for communication between the processor 104 and the system memory 106 .

Depending on the desired configuration, the processor 104 may be any type of process including, but not limited to, a microprocessor (μP), a microcontroller (μC), a digital information processor (DSP), or any combination thereof. Processor 104 may include one or more levels of cache, such as L1 cache 110 and L2 cache 112 , processor core 114 , and registers 116 . Exemplary processor cores 114 may include arithmetic logic units (ALUs), floating point units (FPUs), digital signal processing cores (DSP cores), or any combination thereof. The exemplary memory controller 118 may be used with the processor 104 , or in some implementations, the memory controller 118 may be an internal part of the processor 104 .

Depending on the desired configuration, system memory 106 may be any type of memory including, but not limited to, volatile memory (such as RAM), non-volatile memory (such as ROM, flash memory, etc.), or any combination thereof. System memory 106 may include operating system 120 , one or more applications 122 , and program data 124 . In some embodiments, application 122 may be arranged to operate with program data 124 on an operating system. The program data 124 includes instructions, in the computing device 100 according to the present invention, the program data 124 includes instructions for performing the job image correction method 200 and/or the job topic segmentation method 800 .

Computing device 100 may also include an interface bus 140 that facilitates communication from various interface devices (eg, output device 142 , peripheral interface 144 , and communication device 146 ) to base configuration 102 via bus/interface controller 130 . Example output devices 142 include graphics processing unit 148 and audio processing unit 150 . They may be configured to facilitate communication via one or more A/V ports 152 with various external devices such as displays or speakers. Example peripheral interfaces 144 may include serial interface controller 154 and parallel interface controller 156, which may be configured to facilitate communication via one or more I/O ports 158 and input devices such as keyboard, mouse, pen , voice input devices, touch input devices) or other peripherals (eg printers, scanners, etc.) The example communication device 146 may include a network controller 160 that may be arranged to facilitate communication via one or more communication ports 164 with one or more other computing devices 162 over a network communication link.

A network communication link may be one example of a communication medium. Communication media may typically embody computer readable instructions, data structures, program modules in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media. A "modulated data signal" can be a signal of which one or more of its data sets or whose alterations can be made in such a way as to encode information in the signal. By way of non-limiting example, communication media may include wired media, such as wired or leased line networks, and various wireless media, such as acoustic, radio frequency (RF), microwave, infrared (IR), or other wireless media. The term computer readable medium as used herein may include both storage media and communication media.

Computing device 100 may also include a storage interface bus 134 . Storage interface bus 134 enables communication from storage devices 132 (eg, removable storage 136 and non-removable storage 138 ) to base configuration 102 via bus/interface controller 130 . Operating system 120, applications 122, and at least a portion of data 124 may be stored on removable storage 136 and/or non-removable storage 138, and via the storage interface bus when computing device 100 is powered on or applications 122 are to be executed 134 is loaded into system memory 106 and executed by one or more processors 104 .

Application 122 executes on operating system 120, ie operating system 120 provides various interfaces to operate on hardware devices (eg, storage device 132, output device 142, peripheral interface 144, and communication devices), and at the same time provides application context Managed environment (eg, storage space management and allocation, interrupt handling, process management, etc.). Applications 122 utilize interfaces and environments provided by operating system 120 to control computing device 100 to perform corresponding functions. In some implementations, some applications 122 also provide interfaces. In this way, other applications 122 can call these interfaces to implement functions.

Computing device 100 can be implemented as a server, such as a file server, database server, application server, and WEB server, etc., or as part of a small-sized portable (or mobile) electronic device such as a cellular phone, a personal digital Assistants (PDAs), personal media player devices, wireless web browsing devices, personal headsets, application specific devices, or hybrid devices that may include any of the above. Computing device 100 may also be implemented as a personal computer including desktop computer and notebook computer configurations. In some embodiments, computing device 100 is configured to perform job image correction method 200 and/or job topic segmentation method 800 .

FIG. 2 shows a flowchart of a job image correction method 200 according to one embodiment of the present invention. Method 200 is executed in a computing device, such as computing device 100, to perform warp correction processing on a job image.

As shown in FIG. 2, the method starts at step S210. In step S210, an image of the job to be processed is acquired, and each character connected domain in the image is identified.

Wherein, the job image can be obtained by any method such as photographing or scanning, which is not limited in the present invention. According to an embodiment, before recognizing the connected domains of the characters in the image, it may further include the steps of: converting the job image into a binary image, and performing morphological expansion processing on the binary image, so that adjacent characters in the same line can be connected . Here, the expansion is made so that adjacent texts on the same line can be connected. If there are several blank characters or space bars between two characters, they generally belong to two different character connected domains.

Further, the present invention can also first identify the printed text region in the job image, and identify the text communication domain of the printed text region, where the identified text communication domain is the text communication domain of printed characters. If there is only printed text (such as the title part) in a continuous text, or both printed text and handwritten text (similar to fill-in-the-blank questions, the user has filled in the answer), then the text is a text connection domain. If there is only handwritten text in this segment of continuous text (for example, the user answers a question in a blank area), this segment of text is not recognized as a text-connected domain.

Here, the identification of the printed text area may be preceded by morphological expansion, which will only morphologically expand the printed text area. The recognition of the printed text area can also be morphologically expanded, which will morphologically expand the printed text area and the handwritten text area at the same time. The present invention does not limit the sequence of the steps.

It should be understood that the present invention can train a classification model capable of recognizing the printed text region and the handwritten text region, and the classification model is trained by marking the printed text region and the handwritten text region in multiple images as a training set. . The structure and parameters of the classification model can be set by those skilled in the art as required, which is not limited in the present invention.

Figure 3 is a schematic diagram of the text connection domain in a homework image. The homework image includes the class title at the top and the two-column questions at the bottom. The question has a printed question stem, a handwritten portion of the user's answer, and a question stem. Various picture formats. The title of the class and some question stems are purely printed, and are identified as text-connected fields; some fill-in-the-blank questions are both printed and handwritten, and are also recognized as text-connected fields; while the user's answers in the blank area and the picture part of the question are recognized as Non-text communication domains.

Then, in step S220, one or more text communication domains with similar leftmost abscissas are divided into the same group.

Wherein, that the leftmost abscissa is similar means that the difference between the leftmost abscissas of the text communication domain is smaller than the first percentage of the page width. The first percentage is 8%, mainly to account for errors and indentation, of course not limited to this. Here, the job image can be calibrated by using a coordinate system, for example, the upper left corner of the picture is the origin, the rightward is the X axis, and the downward is the Y axis, of course, it is not limited to this. FIG. 4 is a schematic diagram of a plurality of groups divided according to the text communication domain of FIG. 3 , and is divided into corresponding groups according to the starting position of each text box in the image. The three marked lines running through the job image represent the starting positions of the three groups, and the leftmost abscissa of the text communication field on the lines is the same or similar, and is divided into the same group.

Then, in step S230, it is determined that the job image is a single-column layout or a multi-column layout and a corresponding column area according to the divided one or more groups.

In an implementation manner, if the number of groups is 1, it is determined that the job image is a single-column layout, and the column area corresponding to the single-column layout is the minimum abscissa and maximum abscissa of all the text connection fields in the job image. area in between. The upper and lower boundaries of the column area are the upper and lower boundaries of the job image.

In another implementation manner, if the number of groups is not 1, a rectangular area of each group is generated, and the multiple rectangular areas are deduplicated, and the deduplicated rectangular area is the column area.

Here, the left and right boundaries of the rectangular area are the leftmost and rightmost boundaries of all the text communication domains in the corresponding group, and the upper and lower boundaries are the upper and lower boundaries of the job image. Here, all groups are traversed, and a rectangle is calculated. The top of the rectangle coincides with the top of the picture and the bottom of the rectangle coincides with the bottom of the picture. The left value is the minimum abscissa of the leftmost position of all text connected domains in the group, and the right value is the The maximum value of the abscissa of the rightmost position of all text connected domains in the group. FIG. 5 is a schematic diagram of a rectangular area corresponding to each group in the job image, wherein three grouped rectangles are exemplarily shown.

According to one embodiment, the step of deduplicating the plurality of rectangular areas includes: deleting rectangular areas that overlap more than one rectangular area; for the remaining rectangular areas, deleting rectangular areas contained in other rectangular areas, and removing rectangular areas from From the two intersecting rectangular areas, delete the rectangular area with the smaller width.

The present invention retains all groups that do not overlap with other grouping rectangles, and if there is a grouping with overlapping rectangles, first deletes the groupings that overlap with more than one other grouping rectangle. As shown in Figure 5, the grouping rectangle formed by the title "How much money is left in the third class hour" intersects with the other two grouping rectangles, and will be deleted. This way you can remove some of the impact of centered headings on the layout. At this time, if there are still two intersecting groups, select the grouping rectangle with a larger width and delete the grouping rectangle with a smaller width; if a narrower grouping rectangle is a subset of another wider grouping rectangle, then Delete that narrow grouping rectangle. The narrower three grouping rectangles contained in Figure 6 will be deleted.

After merging, deleting and other operations of each rectangular area to remove duplicates, the final number of rectangular areas is the number of columns of the job image. As shown in Figure 7, there are two rectangular areas that have undergone the deduplication operation, and the job image is a double-column layout, and the corresponding two column areas are the two rectangular areas.

Then, in step S240, for each column area, one or more standard communication domains whose widths meet a predetermined condition are determined, and image correction is performed on the column area based on the text lines of the standard communication domains. Among them, the text line is the text center connection line of the connected domain.

Generally, the standard communication domain is one or more text communication domains whose width reaches a predetermined condition. The width reaching the predetermined condition means that the width is greater than or equal to 75% of the column width, which is of course not limited to this. Then, the transformation formula is calculated according to the text line and horizontal line in the standard connected domain, and the image correction is performed on the column area according to the transformation formula and the prior knowledge of the distortion degree of each text line in the distorted image.

For example, for a two-column picture, you can select one or more text connected domains from the connected domains with a width greater than 75% in the left column area as the standard connected domain, and then use the text lines and horizontal lines of the standard connected domain to define the left column area. Perform image correction. Similarly, image correction can be performed on the area in the right column.

Further, the standard connection domain is the text connection domain whose width reaches the predetermined condition, the uppermost character connection domain and the lowermost character connection domain, and there are two standard connection domains in total. At this time, corresponding transformation formulas are respectively calculated according to the text lines and image horizontal lines of the two standard connectivity domains, and the column area is corrected according to the two transformation formulas obtained by calculation.

As mentioned above, the distortion degree of the job image gradually decreases from the upper and lower ends to the middle, the upper image is distorted downward, and the lower image is distorted upward. Therefore, the present invention selects a text connection from the upper and lower parts that meets the standard and is as close as possible to the two ends of the image. domain as a standard connected domain. Based on the transformation relationship between the horizontal line and the text line of the two connected domains, the job image is corrected layer by layer on the upper and lower sides. Here, some prior knowledge of image distortion changes will also be used.

Further, the present invention can connect three standard communication domains from the text communication domains whose width reaches the predetermined condition, the two uppermost character communication domains, the lowermost character communication domain, and the most central character communication domain. area. After that, the job image is corrected based on the transformation formulas of the three character communication domains.

According to an embodiment of the present invention, after step S240, the method 200 further includes the step of: for a multi-column layout, performing image correction on the transition area between the adjacent column areas according to the image correction results of the adjacent two column areas . According to the transformed point coordinates on the adjacent boundary lines of the adjacent column area, the variant formula is calculated by linear fitting, and finally the entire job is corrected horizontally. For example, the area between the two column areas in Figure 7 is a transition area. Knowing the coordinates before and after correction of each point on the two boundary lines in the middle, according to the original work picture on the same horizontal line The transformation method of two points can be linearly fitted to obtain the modification formula of the transition point between the two points, and then complete the image correction of the entire transition area.

According to an embodiment of the present invention, the method 200 further includes the step of correcting the job image according to the vertical straight line and/or the horizontal straight line.

The step of correcting according to the vertical straight line includes: detecting the vertical straight line in the job image, and performing vertical correction on the job image according to the included angle between the vertical straight line and the vertical line of the image. The vertical straight line is a straight line whose angle with the vertical line of the image is less than a predetermined angle and the height of the straight line is greater than the second percentage of the image height (mostly the column line of the image), or the vertical straight line is the vertical frame of the job image Wire. The central column line of the image is preferred, if there is no column line, the border line is used.

The step of correcting according to the horizontal straight line includes: detecting the horizontal straight line in the job image, and performing horizontal correction on the job image according to the included angle between the horizontal straight line and the horizontal line of the image. The horizontal straight line is a straight line whose included angle with the image horizontal line is smaller than a predetermined angle and whose height is greater than the second percentage of the image height, or the horizontal straight line is a horizontal border line of the job image.

In short, it is to correct the vertical inclination of the job image according to the vertical column line or the image border line, and to correct the horizontal inclination of the job image according to the horizontal column line or the image border line. Optionally, the predetermined angle is 15 degrees, and the second percentage is 50%, which is of course not limited thereto.

The above completes the determination of the layout of the job image and the correction of distortion and inclination. On this basis, it is possible to continue to segment a single question area from the job image for single-question storage.

FIG. 8 shows a flow chart of a method 800 for dividing an assignment topic according to an embodiment of the present invention. Method 800 is performed in a computing device, such as computing device 100 . As shown in FIG. 8, the method starts at step S810.

In step S810, for each column area, the first text connected domain of each line of text in the area is determined to form a vertical connected domain set.

Subsequently, in step S820, the question number line is determined based on the format, content and abscissa position of the first several determined character communication fields in sequence from top to bottom.

Among them, the determination rules of the title line include: the title number is Chinese numerals or Arabic numerals; the character format of the title number at the same level is the same, the numbers are continuous, the abscissa is the same or similar, and the punctuation marks after the title number are the same. In addition, the question number presents a tree structure, and the specific question is in the first-level question number at the bottom, and the question number above the bottom question number generally represents the question type. The line where the picture (such as geometric figures, cartoon characters, etc.) is located is not necessarily the title line. Generally, the line where the picture is located is marked as the title area of the closest question number line above the picture. The title number is mostly on the first character or the second character (for the case where the title number is bracketed), so the first several characters can be the first character, or the first two characters or the first three characters.

The single-question segmentation of the homework image can be understood in conjunction with Figure 9. The single-column area is judged separately, and the single-column area is cut horizontally by the text connected domain to form a group of vertically arranged blocks. The block can be a text block or a picture block. Then identify and process each block from top to bottom. In the process, the question number will be the main method to determine whether it is a new question area. At the same time, the question number analysis and recognition will consider the logic of grading and question number continuity.

Specifically, the block to be eliminated is first obtained. If the area is not a text block, such as a picture block, the text block is added to the set of topic blocks currently being recognized. If the block is a text block, it is determined whether the first several characters of the text contain numbers, and if it does not contain numbers, the text block is also added to the set of topic blocks currently being recognized.

If the text block contains a number, it is determined whether the number is consistent with the format of the title of the first line of the title currently being recognized. On the contrary, if it is inconsistent, it generally represents two situations, one is that the current topic has ended and a new higher-level question type has appeared, and the other is that the block still belongs to the current topic.

Therefore, at this time, it is judged whether there is an upper-level question number. If it exists and the format of the upper-level question number is consistent, the identification of the current title area is ended, which means the end of the current level block, and it is determined that the text block is a new upper-level question. The line of question number of the type, that is, the starting block of the next block of the current level. If there is a superior question number but the format of the superior question number is inconsistent, the text block is added to the set of question blocks currently being recognized. If there is no higher-level question number, it is judged whether there are other-level question numbers, and whether the format is consistent with the other-level question numbers. If they are consistent, it still represents the end of the current topic identification, and this block is the starting block of the other-level title number. If there is no other level title number, the current block is added to the set of title blocks being identified.

The present invention not only relies on the identification of single-digit numbers, but also sets the verification and error-tolerant functions for the title number information according to the above-mentioned preset logical judgment. For example, if a possible question number is found, but it does not conform to the same-level question number rule under the same-question number, the possible question number is judged as a common number instead of question number information.

Subsequently, in step S830, the area of each question is determined according to the ordinate positions of two adjacent question number lines, and each question area is divided for storage.

After confirming the topic area, a single topic in the single-column picture is segmented to obtain a single-question picture, and the single-question picture is stored and distributed.

According to the technical solution of the present invention, the layout layout of the job image can be accurately determined, and the distortion correction and skew correction of the image can be completed, so that the job image can be as close to the original layout as possible. At the same time, each single-question image is accurately extracted for subsequent image recognition, single-question assignment, single-question correction and score summary. The invention has high detection accuracy, small calculation amount, high calculation performance, and can perform real-time and fast large-scale processing.

A9. The method according to any one of A1-A8, wherein the text communication domain is a text communication domain of printed characters. A10. The method according to any one of A1-A9, wherein the leftmost abscissa is close means that the difference between the leftmost abscissas of the text communication domain is less than the first percentage of the page width. A11. The method according to any one of A1-A10, further comprising the step of: detecting a vertical straight line in the job image, and performing vertical correction on the job image according to the angle between the vertical straight line and the image vertical line ; and/or detect a horizontal straight line in the job image, and perform lateral correction on the job image according to the included angle between the horizontal straight line and the horizontal line of the image.

A12. The method according to A11, wherein the vertical straight line is a straight line whose included angle with the vertical line of the image is less than a predetermined angle and whose height is greater than the second percentage of the image height, or the vertical line of the job image A border line; the horizontal straight line is a straight line whose included angle with the horizontal line of the image is less than a predetermined angle and whose height is greater than the second percentage of the image height, or is a horizontal border line of the job image. A13. The method according to A12, wherein the width reaching a predetermined condition means that the width is greater than or equal to 75% of the column width, the first percentage is 8%, and the second percentage is 50%, The predetermined angle is 15 degrees.

A14. The method according to any one of A1-A13, further comprising the step of dividing each topic area in each column area: for each column area, determine the first text connection domain of each line of text in the area ; From top to bottom, determine the title line based on the first several character formats, contents and abscissa positions of the determined respective text connection fields in turn; and determine the area of each title according to the ordinate positions of the adjacent two title lines. , and divide each topic area for storage. A15. The method according to A14, wherein the determination rules for the question number row include: the question number is Chinese numerals or Arabic numerals; the first character format of the question number at the same level is the same, the numbers are continuous, the abscissa is the same or similar, and the title number is after the question number. punctuation marks are the same. A16. The method according to A14, further comprising the step of: marking the row where the picture is located as the title area of the closest question number row above the picture.

The various techniques described herein can be implemented in conjunction with hardware or software, or a combination thereof. Thus, the method and apparatus of the present invention, or certain aspects or portions of the method and apparatus of the present invention, may take the form of an embedded tangible medium, such as a removable hard disk, a USB stick, a floppy disk, a CD-ROM, or any other machine-readable storage medium. in the form of program code (ie, instructions) that, when the program is loaded into a machine, such as a computer, and executed by the machine, the machine becomes an apparatus for practicing the invention.

Where the program code is executed on a programmable computer, the computing device typically includes a processor, a storage medium readable by the processor (including volatile and nonvolatile memory and/or storage elements), at least one input device, and at least one output device. Wherein, the memory is configured to store program codes; the processor is configured to execute the method of the present invention according to the instructions in the program codes stored in the memory.

By way of example and not limitation, readable media include readable storage media and communication media. Readable storage media store information such as computer readable instructions, data structures, program modules or other data. Communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media. Combinations of any of the above are also included within the scope of readable media.

In the specification provided herein, the algorithms and displays are not inherently related to any particular computer, virtual system, or other device. Various general purpose systems may also be used with examples of the present invention. The structure required to construct such a system is apparent from the above description. Furthermore, the present invention is not directed to any particular programming language. It is to be understood that various programming languages may be used to implement the inventions described herein, and that the descriptions of specific languages above are intended to disclose the best mode for carrying out the invention.

In the description provided herein, numerous specific details are set forth. It will be understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.

Similarly, it is to be understood that in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together into a single embodiment, figure, or its description. This disclosure, however, should not be interpreted as reflecting an intention that the invention as claimed requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the Detailed Description are hereby expressly incorporated into this Detailed Description, with each claim standing on its own as a separate embodiment of this invention.

Those skilled in the art will appreciate that the modules or units or components of the apparatus in the examples disclosed herein may be arranged in the apparatus as described in this embodiment, or alternatively may be positioned differently from the apparatus in this example in one or more devices. The modules in the preceding examples may be combined into one module or further divided into sub-modules.

Those skilled in the art will appreciate that the modules in the device in the embodiment can be adaptively changed and arranged in one or more devices different from the embodiment. The modules or units or components in the embodiments may be combined into one module or unit or component, and further they may be divided into multiple sub-modules or sub-units or sub-assemblies. All features disclosed in this specification (including accompanying claims, abstract and drawings) and any method so disclosed may be employed in any combination unless at least some of such features and/or procedures or elements are mutually exclusive. All processes or units of equipment are combined. Each feature disclosed in this specification (including accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.

Furthermore, it will be understood by those skilled in the art that although some of the embodiments described herein include certain features, but not others, included in other embodiments, that combinations of features of different embodiments are intended to be within the scope of the invention within and form different embodiments. For example in the following claims, any of the claimed embodiments can be used in any combination.

Furthermore, some of the described embodiments are described herein as methods or combinations of method elements that can be implemented by a processor of a computer system or by other means for performing the described functions. Thus, a processor having the necessary instructions for implementing the method or method element forms means for implementing the method or method element. Furthermore, an element of an apparatus embodiment described herein is an example of a means for carrying out the function performed by the element for the purpose of carrying out the invention.

As used herein, unless otherwise specified, the use of the ordinal numbers "first," "second," "third," etc. to describe common objects merely refers to different instances of similar objects, and is not intended to imply such The objects being described must have a given order in time, space, ordinal, or in any other way.

While the invention has been described in terms of a limited number of embodiments, those skilled in the art will appreciate, having the benefit of the above description, that other embodiments are conceivable within the scope of the invention thus described. Furthermore, it should be noted that the language used in this specification has been principally selected for readability and teaching purposes, rather than to explain or define the subject matter of the invention. Accordingly, many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the appended claims. This disclosure is intended to be illustrative and not restrictive with regard to the scope of the present invention, which is defined by the appended claims.

Claims (10)

1. A job image correction method suitable for execution in a computing device, the method comprising the steps of: Obtain the job image to be processed, and identify the connected domains of each text in the image; Divide one or more text-connected domains with similar leftmost abscissas into the same group; determining that the job image is a single-column layout or a multi-column layout and a corresponding column area according to the divided one or more groups; and For each column area, one or more standard communication domains whose widths meet a predetermined condition are determined, and image correction is performed on the column area based on the text lines of the standard communication domains. 2. The method of claim 1, further comprising the step of: For a multi-column layout, image correction is performed on the transition area between the adjacent column areas according to the image correction results of two adjacent column areas. 3. The method according to claim 1 or 2, wherein the standard communication domains are two uppermost and lowermost text communication domains among the character communication domains whose widths meet a predetermined condition. 4. The method according to claim 3, wherein the step of performing image correction on the corresponding column area based on the text line of the standard connectivity domain comprises: The corresponding transformation formulas are respectively calculated according to the text lines and the image horizontal lines of the two standard connectivity domains, and the column area is corrected according to the two transformation formulas obtained by calculation. 5. The method according to any one of claims 1-4, wherein, according to the divided one or more groups, determining that the job image is a single-column layout or a multi-column layout, and the corresponding column area Steps include: If the number of groups is 1, it is determined that the job image is a single-column layout, and the corresponding column area is the area between the minimum abscissa and the maximum abscissa of all the text connection domains in the job image; If the number of groups is not 1, a rectangular area of each group is generated, and the multiple rectangular areas are deduplicated, and the deduplicated rectangular area is the column area. 6 . The method of claim 5 , wherein the left and right boundaries of the rectangular area are the leftmost and rightmost boundaries of all text-connected domains in the corresponding group, and the upper and lower boundaries are the upper and lower boundaries of the job image. 7 . 7. The method of claim 5, wherein the step of deduplicating the plurality of rectangular regions comprises: delete rectangular areas that coincide with more than one rectangular area; For the remaining rectangular areas, the rectangular areas contained in other rectangular areas are deleted, and the rectangular area with smaller width is deleted from the intersecting two rectangular areas. 8. The method according to any one of claims 1-7, wherein, before identifying each character connected domain in the image, it further comprises the steps of: Convert the job image into a binary image, and perform morphological expansion processing on the binary image, so that adjacent characters in the same line can be connected. 9. A computing device comprising: at least one processor; and a memory in which program instructions are stored; wherein the processor is configured to perform the method of any one of claims 1-8 according to program instructions stored in the memory. 10. A computer-readable storage medium storing program instructions, the program instructions in the computer-readable storage medium can be read by a computing device, so that the computing device performs the execution of any one of claims 1-8. method described.

Images