A method and system for digital processing of customer files

Abstract

The present invention provides a method and system for digital processing of customer archives, which relates to the technical field of data processing. The method comprises: in response to the customer archive being placed in a preset collection area, triggering a scanning unit to perform a scanning operation; in response to the acquired initial scanning image, performing noise recognition, and in response to updating the initial scanning image based on the recognition result to obtain a current scanning image; in response to receiving a scanning termination signal, triggering the scanning unit to stop the scanning operation, and sorting all the acquired current scanning images to obtain an initial arrangement sequence; performing content recognition on each current scanning image, and determining whether there is an archive continuity relationship between each current scanning image in adjacent positions based on the recognition result; in response to the absence of an archive continuity relationship, performing an abnormal mark based on the initial arrangement sequence, and sending the obtained current arrangement sequence to a user end for viewing. The present invention at least improves the efficiency of archive management.

Description

Technical Field

[0001] The present invention relates to data processing technology, and in particular to a method and system for digital processing of customer files. Background Art

[0002] In the current business environment, the traditional customer file management method mainly relies on paper forms and manual operations. This management method has many technical problems, which seriously restricts the improvement of the work efficiency of grassroots teams and business processing efficiency, and also hinders the optimization of enterprise management.

[0003] First, the storage and management of paper forms are inconvenient. A large number of paper files not only take up space, but are also easily affected by moisture, damaged or lost, resulting in the loss of important information. In addition, the query and retrieval of paper files is also very cumbersome and requires a lot of manpower and time.

[0004] Secondly, manual data entry and paper document transmission are inefficient and prone to errors. In the existing archive management process, a large amount of data needs to be manually entered into the computer system, which is prone to data entry errors or omissions, affecting the accuracy and integrity of the data. At the same time, the transmission of paper documents also consumes a lot of time and resources, and is prone to file loss or delays. Summary of the invention

[0005] Based on the above problems, the present invention is proposed to provide a method and system for digital processing of customer files that overcome the above problems or at least partially solve the above problems.

[0006] According to one aspect of the present invention, a method for digital processing of customer files is provided, comprising the following steps:

[0007] In response to the customer file being placed in the preset collection area, triggering the scanning unit to sequentially perform scanning operations on each file page constituting the customer file;

[0008] In response to obtaining an initial scan image corresponding to any archive page based on the scanning operation, performing noise recognition on the initial scan image, and in response to determining a noise region existing in the initial scan image based on the recognition result, updating the initial scan image to obtain a current scan image corresponding to a current scanning order;

[0009] In response to receiving a scanning termination signal sent by the user terminal, triggering the scanning unit to stop the scanning operation, and sorting all currently scanned images acquired based on the corresponding current scanning orders to obtain an initial arrangement sequence;

[0010] Performing content recognition on each current scanned image in the initial arrangement sequence, and determining whether there is an archival continuity relationship between each current scanned image in adjacent positions based on the recognition results;

[0011] In response to the absence of an archival continuity relationship between any of the current scanned images at adjacent positions, an abnormality mark is performed based on the initial arrangement sequence, and the obtained current arrangement sequence is sent to the user terminal for review.

[0012] Optionally, in the method according to the present invention, the preset acquisition area includes a first acquisition sub-area and a second acquisition sub-area surrounding the first acquisition sub-area;

[0013] In response to the customer file being placed in the preset collection area, the scanning unit is triggered to sequentially perform scanning operations on each file page constituting the customer file, including:

[0014] Acquire a main-level light collection value output by a main-level light sensor disposed in a first collection sub-area corresponding to the preset collection area, and compare the main-level light collection value with the retrieved preset light reference value;

[0015] In response to the primary light collection value being less than the preset light reference value, triggering each secondary light collection value outputted by each secondary light sensor disposed in a second collection sub-area corresponding to the preset collection area to be acquired, and comparing each secondary light collection value with the preset light reference value;

[0016] In response to each secondary light collection value being less than the preset light reference value, it is determined that the customer file is placed in the preset collection area, and the scanning unit is triggered to sequentially perform scanning operations on each file page constituting the customer file.

[0017] Optionally, in the method according to the present invention, triggering the scanning unit to sequentially perform scanning operations on each file page constituting the customer file includes:

[0018] The preset acquisition area is gridded to obtain grid points with the same spacing, and elevations of the grid points are collected to obtain elevation values ​​of the points corresponding to the grid points;

[0019] Acquire a plane elevation value corresponding to a regional plane of the preset acquisition area, and determine elevation values ​​of points respectively corresponding to the plane elevation value as a file elevation group;

[0020] Calculate the mean of the elevation values ​​of each point in the archive elevation group, and calculate the ratio of the obtained elevation mean to the retrieved benchmark elevation value to obtain a real-time elevation ratio;

[0021] Adjusting the reference scanning focal length corresponding to the reference elevation value based on the real-time elevation ratio to obtain the real-time scanning focal length;

[0022] Triggering the scanning unit to perform a scanning operation on each file page constituting the customer file based on the real-time scanning focal length, and establishing a scheduled update task based on the scanning operation;

[0023] In response to the scanning unit completing the scanning of the corresponding preset number of archive pages based on the scheduled update task, the real-time scanning focal length is updated based on a preset increase factor corresponding to the preset number.

[0024] Optionally, in the method according to the present invention, in response to obtaining an initial scanned image corresponding to any archive page based on the scanning operation, performing noise recognition on the initial scanned image, and in response to determining that a noise region exists in the initial scanned image based on the recognition result, updating the initial scanned image to obtain a current scanned image corresponding to the current scanning order, comprises:

[0025] Based on the scanning operation, an initial scan image corresponding to any archive page is obtained, a current scan order corresponding to the initial scan image is determined based on the number of historical scans, and contour recognition is performed on the initial scan image to obtain contours of each image constituting the initial scan image;

[0026] Determining a scan attribute corresponding to the initial scan image based on each image contour corresponding to the same initial scan image, wherein the scan attribute includes a complete attribute and an incomplete attribute;

[0027] In response to the scanning attribute of the initial scanning image being a complete attribute, performing noise recognition on the initial scanning image, and determining whether there is a noise area in the initial scanning image based on the recognition result;

[0028] In response to the existence of a noise area in the initial scanned image, the relative position of the noise between the noise area and each image character in the initial scanned image is obtained, and the initial scanned image is updated based on the relative position of the noise to obtain a current scanned image corresponding to the current scan order.

[0029] Optionally, in the method according to the present invention, determining the scanning attribute corresponding to the initial scanned image based on each image contour corresponding to the same initial scanned image comprises:

[0030] Acquire each image contour located in the initial scanned image, and classify each image contour, determine each image contour located at the edge of the initial scanned image as a peripheral contour group, and determine each image contour located inside the initial scanned image as an internal contour group;

[0031] Performing tilt detection on each image contour in the internal contour group in a corresponding reference direction to obtain the tilt of each contour corresponding to each image contour;

[0032] In response to the fact that the contour inclinations corresponding to the image contours are respectively less than the preset inclinations, distance detection corresponding to the nearest character distances is performed on the image contours located in the peripheral contour group to obtain the nearest distances of the characters corresponding to the image contours;

[0033] Determine any one of the closest distances of each character as the target distance, and calculate the difference between the remaining closest distances of all characters and the target distance, to obtain each distance difference;

[0034] In response to any distance difference being greater than a preset difference, the initial scanned image is determined to have an incomplete attribute.

[0035] Optionally, in the method according to the present invention, performing tilt detection based on a reference direction on each image contour in the internal contour group to obtain each contour tilt corresponding to each image contour comprises:

[0036] Acquire a first contour endpoint and a second contour endpoint respectively corresponding to each image contour in the internal contour group, and determine any contour endpoint as a target endpoint;

[0037] Establishing a first reference line and a second reference line extending in the horizontal direction and the vertical direction respectively with the target endpoint as a starting point, and determining a first contour angle and a second contour angle between the image contour and the first reference line and the second reference line respectively;

[0038] The direction corresponding to the smallest contour angle is determined as the reference direction, and the contour angle corresponding to the reference direction is determined as the contour inclination corresponding to the image contour.

[0039] Optionally, in the method according to the present invention, distance detection corresponding to the nearest character distance is performed on each image contour located in the peripheral contour group to obtain the nearest distance of each character corresponding to each image contour, including:

[0040] Determine the position of each image contour located in the peripheral contour group to obtain an upper contour corresponding to the upper position, a lower contour corresponding to the lower position, a left contour corresponding to the left position, and a right contour corresponding to the right position;

[0041] Based on the extension direction of the upper contour, each image character having a character proximity relationship with the upper contour is determined as an upper character group; based on the extension direction of the lower contour, each image character having a character proximity relationship with the lower contour is determined as a lower character group; based on the extension direction of the left contour, each image character having a character proximity relationship with the left contour is determined as a left character group; based on the extension direction of the right contour, each image character having a character proximity relationship with the right contour is determined as a right character group;

[0042] Respectively performing pixel recognition on each image character in the upper character group, the lower character group, the left character group, and the right character group to obtain pixel points of each character;

[0043] Based on each pixel point corresponding to the same character group, the pixel distance between the pixel point and the corresponding image contour is obtained, and the pixel distance with the smallest corresponding distance is taken as the shortest distance of the character of the corresponding image contour.

[0044] Optionally, in the method according to the present invention, in response to the presence of a noise area in the initial scanned image, obtaining a noise relative position between the noise area and each image character in the initial scanned image, and updating the initial scanned image based on the noise relative position to obtain a current scanned image corresponding to the current scan order, comprising:

[0045] In response to the presence of a noise region in the initial scanned image, determining a noise contour corresponding to the noise region, and acquiring a contour center point corresponding to the noise contour;

[0046] Taking the center point of the contour as a reference magnification point and based on the retrieved preset magnification factor, the noise contour is magnified to obtain a character determination contour surrounding the noise contour;

[0047] Determine an area between the noise contour and the character determination contour as a character determination area, and determine whether there is an image character in the character determination area;

[0048] In response to the presence of any image character in the character determination area, triggering the scanning unit to perform a scanning operation on the file page corresponding to the initial scanned image again, and calling a preset reminder voice for audio output;

[0049] In response to the absence of any image character in the character determination area, the initial scanned image is updated based on a preset image replacement strategy.

[0050] Optionally, in the method according to the present invention, in response to the absence of any image character in the character determination area, updating the initial scanned image based on a preset image replacement strategy includes:

[0051] In response to the absence of any image character in the character determination area, acquiring pixels of the character determination area, and determining the obtained regional pixel value corresponding to the character determination area as a reference pixel value;

[0052] Acquire pixels of the noise area to obtain noise pixels constituting the noise area;

[0053] Pixel replacement is performed on each noise pixel value corresponding to each noise pixel point based on the reference pixel value to obtain the updated initial scan image.

[0054] Optionally, in the method according to the present invention, content recognition is performed on each current scanned image in the initial arrangement sequence, and based on the recognition result, it is determined whether there is an archival continuity relationship between each current scanned image in adjacent positions, including:

[0055] Performing paragraph recognition on each current scanned image in the initial arrangement sequence to obtain each character paragraph arranged vertically corresponding to the same current scanned image;

[0056] Determining each current scan image as a primary scan image in sequence based on the initial arrangement sequence, and determining the current scan image located after the primary scan image as a secondary scan image;

[0057] Determine the last character paragraph in the primary scanned image as the preceding connected paragraph, determine the first character paragraph in the secondary scanned image as the following connected paragraph, and obtain the last image character in the preceding connected paragraph and the first image character in the following connected paragraph;

[0058] Retrieving a preset segmentation character group, wherein the preset segmentation character group includes a colon character, a period character, a semicolon character, and a comma character;

[0059] The preset segmented character group is traversed to determine whether the last image character and the first image character are included in the preset segmented character group, and based on the determination result, it is determined whether there is an archival continuity relationship between the current scanned images at adjacent positions.

[0060] Optionally, in the method according to the present invention, judging whether there is an archival continuity relationship between the current scanned images at adjacent positions based on the determination result includes:

[0061] In response to one of the last image character and the first image character being included in the preset segmented character group, determining a primary complete paragraph including the last image character and a secondary complete paragraph including the first image character in the primary scanned image and the secondary scanned image respectively;

[0062] Performing semantic recognition on the primary complete paragraph and the secondary complete paragraph respectively, and determining whether there is an archival continuity relationship between the primary scanned image and the secondary scanned image based on the recognition result;

[0063] In response to the last image character and the first image character not being included in the preset segmented character group, determining a primary complete paragraph including the last image character and a secondary complete paragraph including the first image character in the primary scanned image and the secondary scanned image respectively, and merging the primary complete paragraph and the secondary complete paragraph to obtain a merged complete paragraph;

[0064] Semantic recognition is performed on the merged complete paragraph, and based on the recognition result, it is determined whether there is an archival continuity relationship between the primary scanned image and the secondary scanned image.

[0065] Optionally, in the method according to the present invention, in response to the absence of an archival continuity relationship between any of the current scanned images at adjacent positions, abnormal marking is performed based on the initial arrangement sequence, and the obtained current arrangement sequence is sent to the user terminal for viewing, including:

[0066] In response to the absence of an archival continuity relationship between any of the current scanned images at adjacent positions, a scan display layer is created, and scan display areas having the same number are established in the scan display layer based on the number of images corresponding to the current scanned images, wherein the scan display areas include a sequence display sub-area and an image display sub-area;

[0067] Filling each current scan order into each order display sub-area located in the scan display layer and filling each current scan image into each image display sub-area located in the scan display layer based on the initial arrangement order;

[0068] Establishing image enclosing frames respectively enclosing each current scanned image that does not have the archive connection relationship in the scan display layer, and forming a frame connection line between two image enclosing frames at adjacent positions, so as to complete the abnormal marking of the initial arrangement sequence;

[0069] The scanned display layer corresponding to the current arrangement sequence is sent to the user end for viewing

[0070] According to another aspect of the present invention, there is provided a customer file digital processing system, comprising:

[0071] A scanning module, configured to trigger the scanning unit to sequentially perform scanning operations on each file page constituting the customer file in response to the customer file being placed in the preset collection area;

[0072] a recognition module configured to, in response to obtaining an initial scan image corresponding to any archive page based on the scanning operation, perform noise recognition on the initial scan image, and in response to determining a noise region existing in the initial scan image based on the recognition result, update the initial scan image to obtain a current scan image corresponding to a current scanning order;

[0073] A sorting module is configured to, in response to receiving a scanning termination signal sent by the user terminal, trigger the scanning unit to stop the scanning operation, and sort all currently scanned images acquired based on the corresponding current scanning orders to obtain an initial arrangement sequence;

[0074] a determination module configured to perform content recognition on each current scanned image in the initial arrangement sequence, and determine whether there is an archival continuity relationship between each current scanned image in adjacent positions based on the recognition result;

[0075] The marking module is configured to perform an abnormal marking based on the initial arrangement sequence in response to the absence of an archival continuity relationship between any of the current scanned images in adjacent positions, and send the obtained current arrangement sequence to the user terminal for viewing.

[0076] According to another aspect of the present invention, there is provided a computing device, comprising: at least one processor; and a memory storing program instructions, wherein the program instructions are configured to be suitable for being executed by the at least one processor, and the program instructions include instructions for executing the above method.

[0077] According to another aspect of the present invention, a readable storage medium storing program instructions is provided. When the program instructions are read and executed by a computing device, the computing device executes the above method.

[0078] According to the content of the present invention, when the customer file is placed in a preset collection area, the present invention can trigger the scanning unit to perform a scanning operation on each file page constituting the customer file, and when there is a noise area in the initial scan image corresponding to any file page, the initial scan image can also be updated to ensure that the current scan image obtained does not have a noise area, thereby eliminating the influence of laughter; in addition, after completing the scanning operation on the customer file, it is possible to further determine whether there is a file continuity relationship between the current scan images obtained. If not, a corresponding abnormal mark is made and sent to the user end for viewing; this not only improves the efficiency of electronic entry of customer files and the efficiency of file management, but also can perform intelligent recognition of order, thereby improving the corresponding user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1A flowchart showing a method for digitalizing customer files according to an embodiment of the present invention is shown;

[0080] Figure 2 A schematic diagram of a scanning display layer according to an embodiment of the present invention;

[0081] Figure 3 A structural block diagram of a customer file digital processing system according to another embodiment of the present invention is shown. DETAILED DESCRIPTION

[0082] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0083] An embodiment of the present invention provides a method for digital processing of customer files, which can be executed in a computing device, wherein the computing device may include a terminal with data processing capabilities, such as a mobile phone or a computer.

[0084] Figure 1 A flowchart of a method for digitalizing customer files according to an embodiment of the present invention is shown, wherein the method comprises the following steps (S101-S105):

[0085] S101 . In response to a customer file being placed in a preset collection area, trigger a scanning unit to sequentially perform scanning operations on each file page constituting the customer file.

[0086] For example, in this embodiment, general customer files are multi-page structures, such as multiple file pages are combined and bound by hinges; therefore, when the user needs to perform intelligent page scanning on the customer file, the user can place the corresponding customer file in a pre-set preset collection area, so that the server triggers the scanning unit to perform corresponding scanning operations on each file page that constitutes the customer file in turn. Here, the scanning unit can be understood as a scanner with, for example, an OCR scanning function, which can identify the corresponding paper document as an electronic document.

[0087] It should be noted that since the preset collection area is set in advance, and in order to realize the scanning operation of the customer file, it is necessary to place the customer file in the corresponding preset collection area, and in order to determine whether the customer file is placed in the preset collection area, a corresponding judgment strategy needs to be adopted.

[0088] Here, in order to set a corresponding judgment strategy to determine whether the customer file is placed in the preset collection area, the preset collection area can be divided first to obtain a corresponding first collection sub-area and a second collection sub-area surrounding the first collection sub-area. Then, the above-mentioned "in response to the customer file being placed in the preset collection area, triggering the scanning unit to sequentially perform scanning operations on each file page constituting the customer file" can also include the following steps:

[0089] Acquire a main-level light collection value output by a main-level light sensor disposed in a first collection sub-area corresponding to the preset collection area, and compare the main-level light collection value with the retrieved preset light reference value;

[0090] In response to the primary light collection value being less than the preset light reference value, triggering each secondary light collection value outputted by each secondary light sensor disposed in a second collection sub-area corresponding to the preset collection area to be acquired, and comparing each secondary light collection value with the preset light reference value;

[0091] In response to each secondary light collection value being less than the preset light reference value, it is determined that the customer file is placed in the preset collection area, and the scanning unit is triggered to sequentially perform scanning operations on each file page constituting the customer file.

[0092] For example, in this embodiment, the area size of the preset collection area can be set according to the size of the customer file. For example, when the size of the customer file is an A4-sized paper, the corresponding customer file can also be set to an A4-sized paper size, so that the preset collection area can fully realize the placement of the customer file; after determining the area size corresponding to the preset collection area, the area position of the preset collection area can be further determined. Here, the preset collection area can generally be selected as a desktop, such as placing a disc-shaped carrying container on the desktop.

[0093] After completing the area size and area position of the preset collection area, the corresponding main-level light sensor can be pre-set in the first collection sub-area and the real-time collection of ambient light can be performed at the same time. The main-level collection light value obtained should be explained that since customer files are generally paper documents with poor light transmittance, if the corresponding customer files are placed in the preset collection area, the customer files will interfere with the ambient light collected by the main-level light sensor. At this time, the main-level collection light value output by the main-level light sensor will be significantly lower than the main-level collection light value collected by the main-level light sensor without placing the customer files in the preset collection area; based on the main-level collection light values ​​respectively collected under two different conditions, the server will compare the main-level collection light value obtained in real time with the retrieved preset light reference value. Here, the preset light reference value can be set by the management personnel, which can specifically correspond to the value corresponding to the ambient light in a normal unobstructed environment;

[0094] When the primary light collection value is less than the preset light reference value, it can indicate that there may be a corresponding object in the corresponding first collection sub-area that blocks the primary light sensor. At this time, each secondary light collection value output by each secondary light sensor in the second collection sub-area set in the preset collection area will be triggered to be acquired. Each secondary light collection value is compared with the preset light reference value again. Here, when it is determined that there is a corresponding blocked object in the first collection sub-area, in order to further determine that the object is a customer file, a corresponding plurality of secondary light sensors can also be set in the second collection sub-area, and the plurality of secondary light sensors can be evenly distributed in the corresponding second collection sub-area.

[0095] When each secondary light collection value is less than the preset light reference value, it indicates that the first collection sub-area and the second collection sub-area are both blocked by corresponding objects at the same time. At this time, it can be determined that the customer file should be placed in the preset collection area, so as to further trigger the scanning unit to perform scanning operations on each file page that constitutes the customer file in turn.

[0096] It should be noted that, in the present embodiment, the judgment strategy can be implemented based on two steps, firstly, based on the primary light sensor of the first collection sub-area, and secondly, based on the secondary light sensors of the second collection sub-area; and each secondary light sensor is generally in a dormant state, that is, no real-time light collection is performed, and when the primary light collection value collected by the primary light sensor is less than the preset light reference value, the server will control the triggering of each secondary light sensor to perform corresponding light collection. Therefore, the above judgment strategy not only realizes the automated processing of the location determination of the customer file, but also improves the accuracy of the location determination of the customer file.

[0097] Furthermore, in this embodiment, the above-mentioned “triggering the scanning unit to sequentially perform scanning operations on each file page constituting the customer file” may also include the following steps:

[0098] The preset acquisition area is gridded to obtain grid points with the same spacing, and elevations of the grid points are collected to obtain elevation values ​​of the points corresponding to the grid points;

[0099] Acquire a plane elevation value corresponding to a regional plane of the preset acquisition area, and determine elevation values ​​of points respectively corresponding to the plane elevation value as a file elevation group;

[0100] Calculate the mean of the elevation values ​​of each point in the archive elevation group, and calculate the ratio of the obtained elevation mean to the retrieved benchmark elevation value to obtain a real-time elevation ratio;

[0101] Adjusting the reference scanning focal length corresponding to the reference elevation value based on the real-time elevation ratio to obtain the real-time scanning focal length;

[0102] Triggering the scanning unit to perform a scanning operation on each file page constituting the customer file based on the real-time scanning focal length, and establishing a scheduled update task based on the scanning operation;

[0103] In response to the scanning unit completing the scanning of the corresponding preset number of archive pages based on the scheduled update task, the real-time scanning focal length is updated based on a preset increase factor corresponding to the preset number.

[0104] For example, in this embodiment, scanning of each file page constituting the customer file can be performed by means of OCR, for example, which specifically includes the need to control the scanning frame of the corresponding scanning unit to scan the customer file, and the corresponding focal length needs to be reasonably set so that the customer file can be completely surrounded by the scanning frame, and the content in the customer file is guaranteed to have a certain clarity; it should be noted that, based on the principle of near big and far small, the smaller the focal length, the smaller the electronic file of the corresponding customer file scanned by the scanning frame, and the lower the clarity of the corresponding content; the larger the focal length, the larger the electronic file of the corresponding customer file scanned by the scanning frame, and the higher the clarity of the corresponding content; and since the customer file has multiple pages of file pages, and each page of the file has a large image, the image is relatively clear. The pages generally have the same paper thickness. When performing the corresponding scanning, when the scanning of a file page is completed, the file page needs to be turned over to leak out other file pages under the file page for re-scanning. After completing the scanning of each file page, the thickness of the remaining unscanned part of the corresponding customer file will become smaller and smaller, which will cause the obtained electronic file to become smaller and smaller. At this time, the electronic file of the corresponding customer file scanned by the scanning frame may have unclear content. Based on this, the focal length of the scanning unit can be dynamically adjusted according to the scanning situation of each file page to ensure that each scanned electronic file has clear content; specifically, the corresponding adjustment process can be as follows:

[0105] First, in order to realize dynamic adjustment of the focal length of the scanning unit, it is necessary to first determine the focal length of the scanning unit in the initial state. Here, the focal length in the initial state can be determined based on the preset collection area where the customer file is placed. Since the surface of the customer file may not be in a horizontal state due to the paper making process, and if the customer file is a plurality of file pages combined and bound by hinges, the corresponding hinges may also affect the determination of the thickness of the customer file, then it is necessary to perform corresponding grid processing on the preset collection area to obtain the corresponding grid points with the same interval, and further obtain the grid points. The corresponding elevation collection is performed on each grid point respectively, so that the elevation value of each point corresponding to each grid point can be obtained. Here, the corresponding elevation collection can be realized by using the corresponding infrared sensor. After the corresponding elevation value of each point is obtained, since the elevation value of each point may have the elevation value of the corresponding customer file and the elevation value of the regional plane corresponding to the preset collection area, it is necessary to obtain the plane elevation value corresponding to the regional plane of the preset collection area, and further determine the elevation value of each point greater than the plane elevation value as the file elevation value, so as to remove the elevation value of the regional plane corresponding to the preset collection area;

[0106] Next, based on the obtained archive elevation group, the elevation values ​​of each point located therein can be averaged, and the corresponding ratio calculation can be performed based on the obtained elevation average and the retrieved benchmark elevation value, so as to obtain the real-time elevation ratio. Here, the benchmark elevation value can be set by the management personnel, which can correspond to a certain distance value between the scanning unit and the preset acquisition area;

[0107] Next, the above-mentioned reference elevation value can correspond to the reference scanning focal length. The reference scanning focal length is the focal length at which the scanning unit can determine to obtain a complete electronic file with clear content when scanning the file page at a fixed height. Based on the principle of near large and far small, the larger the focal length, the larger the corresponding scanned electronic file, and the smaller the focal length, the smaller the corresponding scanned electronic file; therefore, the above-mentioned real-time elevation ratio can be adjusted with the reference scanning focal length corresponding to the reference elevation value, so as to obtain the corresponding real-time scanning focal length;

[0108] Then, after completing the adjustment of the reference scanning focal length, the server can trigger the scanning unit to perform corresponding scanning operations on each file page constituting the customer file based on the obtained real-time scanning focal length. It should be noted that the real-time scanning focal length obtained based on the adjustment of the reference scanning focal length can be regarded as the focal length of the corresponding scanning unit in the initial state.

[0109] Finally, in order to achieve dynamic adjustment of the real-time scanning focal length, a scheduled update task can be established based on the scanning operation; here, the scheduled update task can include a preset increase multiple corresponding to a preset number, that is, after completing the scanning of the corresponding preset number of archive pages, the real-time scanning focal length can be updated in response to the preset increase multiple, thereby ensuring that the electronic files corresponding to each archive page are complete and have clear content.

[0110] S102, in response to obtaining an initial scan image corresponding to any archive page based on the scanning operation, performing noise recognition on the initial scan image, and in response to determining the noise area existing in the initial scan image based on the recognition result, updating the initial scan image to obtain a current scan image corresponding to the current scanning order.

[0111] For example, in this embodiment, after completing the scanning operation for any file page, an initial scan image corresponding to the file page can be obtained. Since the scanning operation for each file page in sequence needs to be completed based on the user's active page turning, on the one hand, during the page turning process, if the user's page turning speed is slow, it may cause the previous file page to not be turned yet, and the scanning unit immediately scans the next file page, resulting in the existence of partial occlusion content corresponding to the previous file page in the obtained initial scan image; on the other hand, if the user's finger is placed on any file page during the scanning operation, it may cause the existence of partial occlusion content corresponding to the user in the obtained initial scan image; the partial occlusion content obtained from the above two aspects will have a certain impact on the scanning result of the file page, and in order to eliminate the impact and obtain a complete and clear initial scan image, it is necessary to perform corresponding noise recognition, and further update the initial scan image based on the recognition result, and after completing the corresponding update, the current scan image corresponding to the current scanning order can be obtained.

[0112] Furthermore, in this embodiment, the above-mentioned “in response to obtaining an initial scan image corresponding to any archive page based on the scanning operation, performing noise recognition on the initial scan image, and in response to determining that there is a noise area in the initial scan image based on the recognition result, updating the initial scan image to obtain a current scan image corresponding to the current scanning order” may also include the following steps:

[0113] Based on the scanning operation, an initial scan image corresponding to any archive page is obtained, a current scan order corresponding to the initial scan image is determined based on the number of historical scans, and contour recognition is performed on the initial scan image to obtain contours of each image constituting the initial scan image;

[0114] Determining a scan attribute corresponding to the initial scan image based on each image contour corresponding to the same initial scan image, wherein the scan attribute includes a complete attribute and an incomplete attribute;

[0115] In response to the scanning attribute of the initial scanning image being a complete attribute, performing noise recognition on the initial scanning image, and determining whether there is a noise area in the initial scanning image based on the recognition result;

[0116] In response to the existence of a noise area in the initial scanned image, the relative position of the noise between the noise area and each image character in the initial scanned image is obtained, and the initial scanned image is updated based on the relative position of the noise to obtain a current scanned image corresponding to the current scan order.

[0117] For example, in this embodiment, after the initial scan image corresponding to any file page is obtained, the current scanning order corresponding to the initial scan image can be determined based on the historical scanning times; from the above content, it can be seen that when the corresponding initial scan image has the aforementioned partial occlusion content, that is, the noise area, a corresponding contour will be generated in the initial scan image, so the initial scan image is subjected to corresponding contour recognition, so as to obtain the contours of each image constituting the initial scan image, and further judge the scanning attribute of the initial scan image based on each image contour, wherein, when there is a noise area in the initial scan image, if the scanning attribute of the initial scan image is a complete attribute, it indicates that the noise area may be generated based on the placement of the finger on the file page, and if the scanning attribute of the initial scan image is an incomplete attribute, it indicates that the noise area may be based on an incomplete Based on these two different scanning attributes, different further updating processes can be performed. For example, when the scanning attribute of the initial scanned image is a complete attribute, corresponding noise recognition can be performed on the initial scanned image (the recognition process can be specifically inputting the initial scanned image into a pre-trained image recognition model) to determine whether there is a noise area in the initial scanned image. When there is a noise area in the initial scanned image, the initial scanned image can be updated accordingly by obtaining the relative position of the noise between the noise area and each image character in the initial scanned image, so as to obtain the current scanned image corresponding to the current scanning order. When the scanning attribute of the initial scanned image is an incomplete attribute, it indicates that it is necessary to perform another scan of the archive page, so as to obtain the initial scanned image corresponding to the complete attribute.

[0118] For example, in this embodiment, the above-mentioned “determining the scanning attribute corresponding to the initial scanned image based on each image contour corresponding to the same initial scanned image” may also include the following steps:

[0119] Acquire each image contour located in the initial scanned image, and classify each image contour, determine each image contour located at the edge of the initial scanned image as a peripheral contour group, and determine each image contour located inside the initial scanned image as an internal contour group;

[0120] Performing tilt detection on each image contour in the internal contour group in a corresponding reference direction to obtain the tilt of each contour corresponding to each image contour;

[0121] In response to the fact that the contour inclinations corresponding to the image contours are respectively less than the preset inclinations, distance detection corresponding to the nearest character distances is performed on the image contours located in the peripheral contour group to obtain the nearest distances of the characters corresponding to the image contours;

[0122] Determine any one of the closest distances of each character as the target distance, and calculate the difference between the remaining closest distances of all characters and the target distance, to obtain each distance difference;

[0123] In response to any distance difference being greater than a preset difference, the initial scanned image is determined to have an incomplete attribute.

[0124] It should be noted that when the noise area is generated based on the incomplete page turning, the image contours of two consecutively arranged archive pages will appear in the same initial scanned image. On the one hand, corresponding tilts may appear in the corresponding image contours. On the other hand, under normal circumstances, each image character in the initial scanned image should have a fixed distance from the edge of the initial scanned image. If the page turning is not completed, image characters close to the edge may appear at the edge of the corresponding initial scanned image. Based on these two characteristics, the scanning properties of the initial scanned image can be judged accordingly. The specific process includes the following:

[0125] First, each image contour in the initial scanned image can be obtained, and each image contour is further classified, so that each image contour located at the edge of the initial scanned image and each image contour located inside the initial scanned image are determined as an internal contour group, where "inside" is obtained relative to "edge";

[0126] Next, based on the two contour groups corresponding to different positions, detections corresponding to different characteristics can be performed. Based on the internal contour group, the image contours located therein can be respectively detected to be tilted in a corresponding reference direction to obtain the tilts of the contours corresponding to the image contours. Here, the reference direction may include a horizontal direction and / or a vertical direction, and the corresponding contour tilt can be understood as the angle formed between the image contour and the reference direction.

[0127] Next, after obtaining the corresponding contour inclination, the contour inclination can be compared with the retrieved preset inclination. When the contour inclination of any image contour is greater than or equal to the preset inclination, it indicates that the image contour has a large degree of inclination. At this time, the scanning attribute of the initial scanned image can be determined as an incomplete attribute; and when the inclinations of each contour are less than the preset inclination, the distance detection corresponding to the nearest character distance can be further performed on each image contour located in the peripheral contour group, thereby obtaining the nearest distance of each character corresponding to each image contour;

[0128] Then, the corresponding minimum distance among the closest distances of each character is determined as the target distance, and the difference between the target distance and the closest distances of all the remaining characters is calculated in sequence to obtain the corresponding distance differences;

[0129] Finally, the distance difference values ​​obtained above are compared with the preset difference values ​​respectively. When any distance difference value is greater than the preset difference value, it can be indicated that the scanning attribute of the current initial scanned image is an incomplete attribute.

[0130] It should be noted that in this embodiment, the file format in any file page should be consistent under normal circumstances. Therefore, there should be a relatively fixed spacing distance between each image character and the corresponding image contour located in the same file page. Therefore, if any distance difference is greater than the preset difference, image characters close to the edge may appear at the edge of the corresponding initial scanned image; similarly, each image contour located in the same file page should also correspond to the reference direction. Therefore, based on the above judgment process, the scanning properties of the initial scanned image can be accurately judged, so as to further perform different update processes based on different scanning properties.

[0131] Furthermore, in this embodiment, the above-mentioned “performing tilt detection on each image contour in the internal contour group based on the reference direction to obtain each contour tilt corresponding to each image contour” may also include the following steps:

[0132] Acquire a first contour endpoint and a second contour endpoint respectively corresponding to each image contour in the internal contour group, and determine any contour endpoint as a target endpoint;

[0133] Establishing a first reference line and a second reference line extending in the horizontal direction and the vertical direction respectively with the target endpoint as a starting point, and determining a first contour angle and a second contour angle between the image contour and the first reference line and the second reference line respectively;

[0134] The direction corresponding to the smallest contour angle is determined as the reference direction, and the contour angle corresponding to the reference direction is determined as the contour inclination corresponding to the image contour.

[0135] For example, in the present embodiment, when performing tilt detection based on the reference direction on each image contour, the first contour endpoint and the second contour endpoint corresponding to each image contour in the internal contour group can be first obtained, so as to further determine any contour endpoint as the corresponding target endpoint; further, since the image contour of a general archive page generally includes contour lines extending in the horizontal direction and in the vertical direction, the first reference line and the second reference line extending in the horizontal direction and in the vertical direction can be established with the target endpoint as the starting point, and then the first contour angle and the second contour angle between the image contour and the established first reference line and second reference line can be determined; finally, the direction with the smallest corresponding contour angle can be determined as the reference direction, and the contour angle corresponding to the reference direction can be further determined as the contour inclination corresponding to the image contour.

[0136] Similarly, in this embodiment, the above-mentioned “performing distance detection for each image contour located in the outer contour group corresponding to the nearest character distance, and obtaining the nearest distance of each character corresponding to each image contour” may also include the following steps:

[0137] Determine the position of each image contour located in the peripheral contour group to obtain an upper contour corresponding to the upper position, a lower contour corresponding to the lower position, a left contour corresponding to the left position, and a right contour corresponding to the right position;

[0138] Based on the extension direction of the upper contour, each image character having a character proximity relationship with the upper contour is determined as an upper character group; based on the extension direction of the lower contour, each image character having a character proximity relationship with the lower contour is determined as a lower character group; based on the extension direction of the left contour, each image character having a character proximity relationship with the left contour is determined as a left character group; based on the extension direction of the right contour, each image character having a character proximity relationship with the right contour is determined as a right character group;

[0139] Respectively performing pixel recognition on each image character in the upper character group, the lower character group, the left character group, and the right character group to obtain pixel points of each character;

[0140] Based on each pixel point corresponding to the same character group, the pixel distance between the pixel point and the corresponding image contour is obtained, and the pixel distance with the smallest corresponding distance is taken as the shortest distance of the character of the corresponding image contour.

[0141] For example, in the present embodiment, since each image contour located in the outer contour group is generally an image contour that surrounds the entire image character, for example, each image contour can form a corresponding rectangular shape, therefore, the position can be determined based on each image contour located in the outer contour group to obtain an upper contour corresponding to the upper position, a lower contour corresponding to the lower position, a left contour corresponding to the left position, and a right contour corresponding to the right position; after the contours at different positions are obtained, each image character having a character proximity relationship with the contour can be determined as a corresponding character group based on the extension direction of each contour, that is, the upper character group, the lower character group, the left character group, and the right character group mentioned above; finally, each image character located in each character group can be pixel-recognized to obtain each character pixel point, and further based on each pixel point corresponding to the same character group, the pixel point between the corresponding image contour and the pixel point can be obtained, so as to determine the pixel distance with the smallest corresponding distance as the character closest distance of the corresponding image contour; it should be noted that, in the present embodiment, the above-mentioned "character proximity relationship" can be understood as when there is no other character between any image character and the image contour, the two can be determined as having a corresponding "character proximity relationship".

[0142] In this embodiment, based on the above content, it can be known that when the scanning attribute of the initial scanned image is a complete attribute, the initial scanned image can be subjected to corresponding noise recognition, and correspondingly, the above-mentioned "in response to the existence of a noise region in the initial scanned image, obtaining the noise relative position between the noise region and each image character in the initial scanned image, and updating the initial scanned image based on the noise relative position to obtain a current scanned image corresponding to the current scanning order" can be achieved through the following steps:

[0143] In response to the presence of a noise region in the initial scanned image, determining a noise contour corresponding to the noise region, and acquiring a contour center point corresponding to the noise contour;

[0144] Taking the center point of the contour as a reference magnification point and based on the retrieved preset magnification factor, the noise contour is magnified to obtain a character determination contour surrounding the noise contour;

[0145] Determine an area between the noise contour and the character determination contour as a character determination area, and determine whether there is an image character in the character determination area;

[0146] In response to the presence of any image character in the character determination area, triggering the scanning unit to perform a scanning operation on the file page corresponding to the initial scanned image again, and calling a preset reminder voice for audio output;

[0147] In response to the absence of any image character in the character determination area, the initial scanned image is updated based on a preset image replacement strategy.

[0148] It should be noted that, under the complete attribute, if there is a noise area in the initial scanned image, it means that the noise area may be generated based on the finger placed on the file page. At this time, the following update process can be performed:

[0149] First, after the noise area is determined in the initial scanned image, it is necessary to determine the noise contour corresponding to the noise area, and further obtain the contour center point corresponding to the noise contour;

[0150] Next, in order to determine the relative position of the noise between the noise area and each image character, it is necessary to determine the image characters located around the noise area. Specifically, the center point of the contour can be used as the reference magnification point, and the noise contour can be magnified based on the preset magnification factor, so as to obtain a character determination contour that can surround the noise contour, and determine the area between the noise contour and the character determination contour as the character determination area. In order to further determine whether there is a corresponding image character in the character determination area;

[0151] Finally, when there is any image character in the character determination area, it indicates that corresponding image characters appear around the noise area. In this case, it can be determined that the noise area may have an occlusion relationship with some image characters. At this time, the initial scanned image needs to be scanned again. The specific process is to trigger the scanning unit to perform the scanning operation of the file page corresponding to the initial scanned image again, and it is necessary to call the preset reminder voice for voice output to inform the user; here, the preset reminder voice may include "A noise area that occludes the image characters is detected, please scan again", and the corresponding voice output can be output through the audio unit. Specifically, the audio unit can be integrated with the scanning unit; and when there is no image character in the character determination area, it indicates that there may be no corresponding occlusion relationship between the noise area and the image characters. At this time, the initial scanned image can be updated based on the preset image replacement strategy, and the removal of the corresponding noise area can be completed without scanning again.

[0152] Furthermore, in this embodiment, the above-mentioned “in response to the absence of any image character in the character determination area, updating the initial scanned image based on a preset image replacement strategy” may also include the following steps:

[0153] In response to the absence of any image character in the character determination area, acquiring pixels of the character determination area, and determining the obtained regional pixel value corresponding to the character determination area as a reference pixel value;

[0154] Acquire pixels of the noise area to obtain noise pixels constituting the noise area;

[0155] Pixel replacement is performed on each noise pixel value corresponding to each noise pixel point based on the reference pixel value to obtain the updated initial scan image.

[0156] For example, in this embodiment, based on the above content, it can be known that when there is no image character in the character determination area, it indicates that there may be no corresponding occlusion relationship between the noise area and the image character. At this time, the corresponding area of ​​the noise area can be directly replaced to complete the corresponding update process. Specifically, pixels of the character determination area can be extracted, and the obtained area pixel value corresponding to the character determination area is determined as the reference pixel value. At the same time, pixels of the noise area can be acquired to obtain the noise pixel points that constitute the noise area, so as to replace the noise pixel points based on the reference pixel value to obtain an updated initial scanned image.

[0157] S103 , in response to receiving a scanning termination signal sent by the user terminal, triggering the scanning unit to stop the scanning operation, and sorting all currently scanned images acquired based on the corresponding current scanning orders to obtain an initial arrangement sequence.

[0158] For example, in the present embodiment, since the scanning of the archive volume is performed based on the user manually turning pages, after completing the scanning of each archive page in the archive volume, the user terminal may send a corresponding scanning termination signal to confirm the completion of the corresponding scanning operation. At this time, the server will control the trigger scanning unit to stop the corresponding scanning operation, and further sort all the currently scanned images corresponding to all the archive pages respectively acquired based on the current scanning order, thereby obtaining the corresponding initial arrangement sequence.

[0159] S104, performing content recognition on each current scanned image in the initial arrangement sequence, and determining whether there is an archive continuity relationship between each current scanned image in adjacent positions based on the recognition results.

[0160] Here, since the archive volume is based on hinges to merge and bind the archive pages, if the contents between the archive pages do not have an archive continuity relationship during the merging and binding process, it will affect the user's reading later. Therefore, corresponding content recognition can be performed based on the current scanned images obtained, so as to determine whether there is an archive continuity relationship between the current scanned images in adjacent positions based on the recognition results, so as to determine whether corresponding adjustments need to be made.

[0161] Furthermore, in this embodiment, the above-mentioned “respectively performing content recognition on each current scanned image in the initial arrangement sequence, and respectively determining whether there is an archival continuity relationship between each current scanned image at adjacent positions based on the recognition result” may also include the following steps:

[0162] Performing paragraph recognition on each current scanned image in the initial arrangement sequence to obtain each character paragraph arranged vertically corresponding to the same current scanned image;

[0163] Determining each current scan image as a primary scan image in sequence based on the initial arrangement sequence, and determining the current scan image located after the primary scan image as a secondary scan image;

[0164] Determine the last character paragraph in the primary scanned image as the preceding connected paragraph, determine the first character paragraph in the secondary scanned image as the following connected paragraph, and obtain the last image character in the preceding connected paragraph and the first image character in the following connected paragraph;

[0165] Retrieving a preset segmentation character group, wherein the preset segmentation character group includes a colon character, a period character, a semicolon character, and a comma character;

[0166] The preset segmented character group is traversed to determine whether the last image character and the first image character are included in the preset segmented character group, and based on the determination result, it is determined whether there is an archival continuity relationship between the current scanned images at adjacent positions.

[0167] For example, in this embodiment, determining whether there is an archival continuity relationship between the current scanned images at adjacent positions may specifically include the following process:

[0168] First, paragraph recognition can be performed on each current scanned image based on a pre-trained paragraph recognition model to obtain each character paragraph arranged vertically corresponding to the same current scanned image;

[0169] Next, each current scanned image may be determined as a primary scanned image in sequence based on the corresponding initial arrangement sequence. For example, when there are five current scanned images determined to be included in the initial arrangement sequence, the five current scanned images may be determined as primary scanned images in sequence, and for any primary scanned image, the current scanned image located after it also needs to be determined as a secondary scanned image.

[0170] Next, since the primary scanned image and the secondary scanned image have a corresponding adjacent position relationship, the character paragraph at the end of the primary scanned image can be determined as the corresponding preceding connected paragraph, and the character paragraph at the beginning of the secondary scanned image can be determined as the following connected paragraph. Based on this, the last image character in the preceding connected paragraph and the first image character in the following connected paragraph can be further obtained according to the preceding connected paragraph and the following connected paragraph.

[0171] Finally, the preset segmentation character group can be retrieved from the server, wherein the preset segmentation character group can include a colon character (also known as “:”), a period character (also known as “.”), and a semicolon character (also known as “;”); here, according to general paragraph writing habits, semantically complete paragraphs are generally divided into corresponding paragraphs using the above-mentioned colon characters, period characters, semicolon characters, and comma characters. Therefore, based on this feature, the preset segmentation character group can be traversed to determine whether the last image character and the first image character are included in the preset segmentation character group, and based on the determination result, it can be determined whether there is an archival continuity relationship between the current scanned images in adjacent positions.

[0172] It should be noted that, based on the above characteristics, it can be known that when the last image character and the first image character are both included in the corresponding preset segmented character group, a continuously arranged colon character, a period character, a semicolon character and a comma character or a randomly arranged continuous character can be formed. In this case, it can be directly determined that there is no corresponding file continuity relationship between the corresponding primary scan image and the secondary scan image.

[0173] Furthermore, in this embodiment, the above-mentioned “determining whether there is an archive continuity relationship between the current scanned images at adjacent positions based on the determination result” may also include the following steps:

[0174] In response to one of the last image character and the first image character being included in the preset segmented character group, determining a primary complete paragraph including the last image character and a secondary complete paragraph including the first image character in the primary scanned image and the secondary scanned image respectively;

[0175] Performing semantic recognition on the primary complete paragraph and the secondary complete paragraph respectively, and determining whether there is an archival continuity relationship between the primary scanned image and the secondary scanned image based on the recognition result;

[0176] In response to the last image character and the first image character not being included in the preset segmented character group, determining a primary complete paragraph including the last image character and a secondary complete paragraph including the first image character in the primary scanned image and the secondary scanned image respectively, and merging the primary complete paragraph and the secondary complete paragraph to obtain a merged complete paragraph;

[0177] Semantic recognition is performed on the merged complete paragraph, and based on the recognition result, it is determined whether there is an archival continuity relationship between the primary scanned image and the secondary scanned image.

[0178] For example, in this embodiment, after completing the traversal of the preset segmented character group, whether there is an archival continuity relationship between the primary scanned image and the secondary scanned image may be further determined based on the following different situations:

[0179] The first one is that when one of the last image character and the first image character is included in the preset segmented character group, it indicates that the last image character and the first image character are divided into different paragraphs. If the archival continuity relationship between the main scanned image and the secondary scanned image is to be determined, the main complete paragraph containing the last image character and the secondary complete paragraph containing the first image character need to be determined in the main scanned image and the secondary scanned image respectively, and further a pre-trained semantic recognition model is used to perform semantic recognition on the main complete paragraph and the secondary complete paragraph respectively to obtain corresponding recognition results. If the semantics of the two are close, the corresponding main scanned image and the secondary scanned image can be determined as having an archival continuity relationship. For example, if the semantics of the main complete paragraph include an introduction to the setting function of the photovoltaic panel, and the semantics of the secondary complete paragraph include an introduction to the setting conditions of the photovoltaic panel, then it can be considered that there is an archival continuity relationship between the two.

[0180] The second type is that when the last image character and the first image character are not included in the preset segmented character group, it means that the last image character and the first image character are image characters of corresponding texts arranged continuously, and the two belong to the same paragraph. At this time, it is necessary to determine the main complete paragraph of the last image character and the secondary complete paragraph of the first image character in the main scan image and the secondary scan image respectively, and merge the main complete paragraph and the secondary complete paragraph to obtain a merged complete paragraph; by performing semantic recognition on the merged complete paragraph, it is determined whether there are corresponding logical errors or grammatical errors therein. If not, the corresponding main scan image and the secondary scan image can be determined as having an archival continuity relationship; for example, the main complete paragraph includes "the structure of the photovoltaic panel is large", and the secondary complete paragraph includes "and the weight is heavy", and then it can be seen that the merged complete paragraph position "the structure of the photovoltaic panel is large and the weight is heavy" after the merger does not have any logical errors or grammatical errors, so it can be considered that there is an archival continuity relationship between the two.

[0181] S105 , in response to the absence of a file continuity relationship between any of the current scanned images in adjacent positions, abnormal marking is performed based on the initial arrangement sequence, and the obtained current arrangement sequence is sent to the user terminal for viewing.

[0182] For example, in this embodiment, when there is no file continuity relationship between the current scanned images in adjacent positions, it indicates that the placement order of the two is wrong, then it is necessary to mark the initial arrangement sequence as abnormal, and send the obtained current arrangement sequence to the user end for viewing, so as to make corresponding adjustments.

[0183] Furthermore, in this embodiment, the above-mentioned “in response to the absence of an archival continuity relationship between any of the current scanned images at adjacent positions, abnormal marking is performed based on the initial arrangement sequence, and the obtained current arrangement sequence is sent to the user terminal for viewing” may also include the following steps:

[0184] In response to the absence of an archival continuity relationship between any of the current scanned images at adjacent positions, a scan display layer is created, and scan display areas having the same number are established in the scan display layer based on the number of images corresponding to the current scanned images, wherein the scan display areas include a sequence display sub-area and an image display sub-area;

[0185] Filling each current scan order into each order display sub-area located in the scan display layer and filling each current scan image into each image display sub-area located in the scan display layer based on the initial arrangement order;

[0186] Establishing image enclosing frames respectively enclosing each current scanned image that does not have the archive connection relationship in the scan display layer, and forming a frame connection line between two image enclosing frames at adjacent positions, so as to complete the abnormal marking of the initial arrangement sequence;

[0187] The obtained scanned display layer corresponding to the current arrangement sequence is sent to the user terminal for viewing.

[0188] For example, in this embodiment, when there is no file continuity relationship between any of the current scanned images in adjacent positions, it indicates that the placement order of the corresponding file pages may be wrong. At this time, the server will create a corresponding scan display layer, and establish scan display areas in the scan display layer with the same number of images as the current scanned images, and then fill the current scanned sequences into the sequence display sub-areas in the scan display layer based on the initial arrangement order, and fill the current scanned images into the image display sub-areas in the scan display layer;

[0189] After the order display sub-area and the image display sub-area are filled, image enclosing frames that respectively enclose the current scanned images that do not have a file connection relationship can be further established in the scan display layer, and corresponding frame connection lines can be further formed between two image enclosing frames at adjacent positions, thereby completing the abnormal marking of the initial arrangement sequence;

[0190] Finally, after the abnormal marking of the initial arrangement sequence is completed, the scanned display layer corresponding to the current arrangement sequence can be sent to the user end for corresponding viewing.

[0191] For example, Figure 2 A schematic diagram of a scan display layer is shown, wherein the layer includes five corresponding scan display areas, and an abnormal mark is located between the first and second scan display areas, including image enclosing frames ( Figure 2 ) and the corresponding frame connection lines ( Figure 2 three horizontal lines between the two dashed boxes in the figure).

[0192] In summary, this embodiment can trigger the scanning unit to perform scanning operations on each file page constituting the customer file when the customer file is placed in the preset collection area, and when there is a noise area in the initial scan image corresponding to any file page, the initial scan image can also be updated to ensure that the current scan image obtained does not have a noise area, thereby eliminating the influence of laughter; in addition, after completing the scanning operation on the customer file, it is possible to further determine whether there is a file continuity relationship between the current scan images obtained. If not, a corresponding abnormal mark is made and sent to the user end for viewing; this not only improves the efficiency of electronic entry of customer files and the efficiency of file management, but also can perform intelligent recognition of order, thereby improving the corresponding user experience.

[0193] Another embodiment of the present invention provides a customer file digital processing system. Figure 3 As the corresponding system block diagram, the system includes:

[0194] A scanning module, configured to trigger the scanning unit to sequentially perform scanning operations on each file page constituting the customer file in response to the customer file being placed in the preset collection area;

[0195] a recognition module configured to, in response to obtaining an initial scan image corresponding to any archive page based on the scanning operation, perform noise recognition on the initial scan image, and in response to determining a noise region existing in the initial scan image based on the recognition result, update the initial scan image to obtain a current scan image corresponding to a current scanning order;

[0196] A sorting module is configured to, in response to receiving a scanning termination signal sent by the user terminal, trigger the scanning unit to stop the scanning operation, and sort all currently scanned images acquired based on the corresponding current scanning orders to obtain an initial arrangement sequence;

[0197] a determination module configured to perform content recognition on each current scanned image in the initial arrangement sequence, and determine whether there is an archival continuity relationship between each current scanned image in adjacent positions based on the recognition result;

[0198] The marking module is configured to perform an abnormal marking based on the initial arrangement sequence in response to the absence of an archival continuity relationship between any of the current scanned images in adjacent positions, and send the obtained current arrangement sequence to the user terminal for viewing.

[0199] According to another aspect of the present invention, there is provided a computing device, comprising: at least one processor; and a memory storing program instructions, wherein the program instructions are configured to be suitable for being executed by the at least one processor, and the program instructions include instructions for executing the above method.

[0200] According to another aspect of the present invention, a readable storage medium storing program instructions is provided. When the program instructions are read and executed by a computing device, the computing device executes the above method.

[0201] In the description provided herein, algorithms and displays are not inherently related to any particular computer, virtual system or other device. Various general purpose systems can also be used together with the examples of the present invention. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the present invention is not directed to any specific programming language either. It should be understood that various programming languages ​​can be utilized to implement the content of the present invention described herein, and the above description of specific languages ​​is for the purpose of disclosing the preferred embodiment of the present invention.

[0202] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0203] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof.

[0204] Those skilled in the art will appreciate that the modules or units or components of the devices in the examples disclosed herein may be arranged in the devices described in the embodiment, or alternatively may be located in one or more devices different from the devices in the examples. The modules in the foregoing examples may be combined into one module or may be divided into multiple submodules.

[0205] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and furthermore may be divided into a plurality of submodules or subunits or subcomponents.

[0206] Furthermore, those skilled in the art will appreciate that although some embodiments described herein include certain features included in other embodiments but not other features, the combination of features from different embodiments is meant to be within the scope of the present invention and to form different embodiments.

[0207] In addition, some of the embodiments are described herein as a combination of methods or method elements that can be implemented by a processor of a computer system or by other devices that perform the functions. Therefore, a processor with necessary instructions for implementing the method or method elements forms a device for implementing the method or method elements. In addition, the elements described herein of the device embodiments are examples of devices that are used to implement the functions performed by the elements for implementing the purpose of the invention.

[0208] As used herein, unless otherwise specified, the use of ordinal numbers "first," "second," "third," etc. to describe common objects merely indicates that different instances of similar objects are involved, and is not intended to imply that the objects so described must have a given order in time, space, order, or in any other manner.

[0209] Although the present invention has been described according to a limited number of embodiments, it will be apparent to those skilled in the art, with the benefit of the above description, that other embodiments may be envisioned within the scope of the invention thus described. In addition, it should be noted that the language used in this specification is primarily selected for readability and instructional purposes, rather than for explaining or limiting the subject matter of the present invention.

Claims

1. A method for digital processing of customer files, characterized in that: The following steps are involved: Acquire a main-level light collection value output by a main-level light sensor disposed in a first collection sub-area corresponding to the preset collection area, and compare the main-level light collection value with the retrieved preset light reference value; In response to the primary light collection value being less than the preset light reference value, triggering each secondary light collection value outputted by each secondary light sensor in a second collection sub-area surrounding the first collection sub-area corresponding to the preset collection area to be acquired, and comparing each secondary light collection value with the preset light reference value; In response to each secondary light collection value being less than a preset light reference value, determining that the customer file is placed in a preset collection area, and triggering the scanning unit to sequentially perform scanning operations on each file page constituting the customer file; In response to obtaining an initial scan image corresponding to any archive page based on the scanning operation, performing noise recognition on the initial scan image, and in response to determining a noise region existing in the initial scan image based on the recognition result, updating the initial scan image to obtain a current scan image corresponding to a current scanning order; In response to receiving a scanning termination signal sent by the user terminal, triggering the scanning unit to stop the scanning operation, and sorting all currently scanned images acquired based on the corresponding current scanning orders to obtain an initial arrangement sequence; Performing content recognition on each current scanned image in the initial arrangement sequence, and determining whether there is an archival continuity relationship between each current scanned image in adjacent positions based on the recognition results; In response to the absence of an archival continuity relationship between any of the current scanned images at adjacent positions, an abnormality mark is performed based on the initial arrangement sequence, and the obtained current arrangement sequence is sent to the user terminal for review.

2. The method for digitalizing customer files according to claim 1, characterized in that: The triggering scanning unit sequentially performs scanning operations on each file page constituting the customer file, including: The preset acquisition area is gridded to obtain grid points with the same spacing, and elevations of the grid points are collected to obtain elevation values ​​of the points corresponding to the grid points; Acquire a plane elevation value corresponding to a regional plane of the preset acquisition area, and determine elevation values ​​of points respectively corresponding to the plane elevation value as a file elevation group; Calculate the mean of the elevation values ​​of each point in the archive elevation group, and calculate the ratio of the obtained elevation mean to the retrieved benchmark elevation value to obtain a real-time elevation ratio; Adjusting a reference scanning focal length corresponding to the reference elevation value based on the real-time elevation ratio to obtain a real-time scanning focal length; Triggering the scanning unit to perform a scanning operation on each file page constituting the customer file based on the real-time scanning focal length, and establishing a scheduled update task based on the scanning operation; In response to the scanning unit completing the scanning of the corresponding preset number of archive pages based on the scheduled update task, the real-time scanning focal length is updated based on a preset increase factor corresponding to the preset number.

3. The method for digitalizing customer files according to claim 1, characterized in that: In response to obtaining an initial scan image corresponding to any archive page based on the scanning operation, performing noise recognition on the initial scan image, and in response to determining that a noise region exists in the initial scan image based on the recognition result, updating the initial scan image to obtain a current scan image corresponding to the current scanning order, including: Based on the scanning operation, an initial scan image corresponding to any archive page is obtained, a current scan order corresponding to the initial scan image is determined based on the number of historical scans, and contour recognition is performed on the initial scan image to obtain contours of each image constituting the initial scan image; Determining a scan attribute corresponding to the initial scan image based on each image contour corresponding to the same initial scan image, wherein the scan attribute includes a complete attribute and an incomplete attribute; In response to the scanning attribute of the initial scanning image being a complete attribute, performing noise recognition on the initial scanning image, and determining whether there is a noise area in the initial scanning image based on the recognition result; In response to the existence of a noise area in the initial scanned image, the relative position of the noise between the noise area and each image character in the initial scanned image is obtained, and the initial scanned image is updated based on the relative position of the noise to obtain a current scanned image corresponding to the current scan order.

4. The method for digitalizing customer files according to claim 3, characterized in that: Determining a scanning attribute corresponding to the initial scanned image based on each image contour corresponding to the same initial scanned image includes: Acquire each image contour located in the initial scanned image, and classify each image contour, determine each image contour located at the edge of the initial scanned image as a peripheral contour group, and determine each image contour located inside the initial scanned image as an internal contour group; Performing tilt detection on each image contour in the internal contour group in a corresponding reference direction to obtain the tilt of each contour corresponding to each image contour; In response to the fact that the contour inclinations corresponding to the image contours are respectively less than the preset inclinations, distance detection corresponding to the nearest character distances is performed on the image contours located in the peripheral contour group to obtain the nearest distances of the characters corresponding to the image contours; Determine any one of the closest distances of each character as the target distance, and calculate the difference between the remaining closest distances of all characters and the target distance, to obtain each distance difference; In response to any distance difference being greater than a preset difference, the initial scanned image is determined to have an incomplete attribute.

5. The method for digitalizing customer files according to claim 4, characterized in that: Performing tilt detection based on a reference direction on each image contour in the internal contour group to obtain each contour tilt corresponding to each image contour, including: Acquire a first contour endpoint and a second contour endpoint respectively corresponding to each image contour in the internal contour group, and determine any contour endpoint as a target endpoint; Establishing a first reference line and a second reference line extending in the horizontal direction and the vertical direction respectively with the target endpoint as a starting point, and determining a first contour angle and a second contour angle between the image contour and the first reference line and the second reference line respectively; The direction corresponding to the smallest contour angle is determined as the reference direction, and the contour angle corresponding to the reference direction is determined as the contour inclination corresponding to the image contour.

6. The method for digitalizing customer files according to claim 4, characterized in that: Performing distance detection on each image contour located in the outer contour group and corresponding to the nearest character distance to obtain the nearest distance of each character corresponding to each image contour, including: Determine the position of each image contour located in the peripheral contour group to obtain an upper contour corresponding to the upper position, a lower contour corresponding to the lower position, a left contour corresponding to the left position, and a right contour corresponding to the right position; Based on the extension direction of the upper contour, each image character having a character proximity relationship with the upper contour is determined as an upper character group; based on the extension direction of the lower contour, each image character having a character proximity relationship with the lower contour is determined as a lower character group; based on the extension direction of the left contour, each image character having a character proximity relationship with the left contour is determined as a left character group; based on the extension direction of the right contour, each image character having a character proximity relationship with the right contour is determined as a right character group; Respectively performing pixel recognition on each image character in the upper character group, the lower character group, the left character group, and the right character group to obtain pixel points of each character; Based on each pixel point corresponding to the same character group, the pixel distance between the corresponding image contour is obtained, and the pixel distance with the smallest corresponding distance is taken as the closest distance of the character of the corresponding image contour.

7. The method for digitalizing customer files according to claim 6, characterized in that: In response to the presence of a noise region in the initial scanned image, obtaining a relative position of noise between the noise region and each image character in the initial scanned image, and updating the initial scanned image based on the relative position of noise to obtain a current scanned image corresponding to a current scan order, including: In response to the presence of a noise area in the initial scanned image, determining a noise contour corresponding to the noise area, and acquiring a contour center point corresponding to the noise contour; Taking the center point of the contour as a reference magnification point and based on the retrieved preset magnification factor, the noise contour is magnified to obtain a character determination contour surrounding the noise contour; Determine an area between the noise contour and the character determination contour as a character determination area, and determine whether there is an image character in the character determination area; In response to the presence of any image character in the character determination area, triggering the scanning unit to perform a scanning operation on the file page corresponding to the initial scanned image again, and calling a preset reminder voice for audio output; In response to the absence of any image character in the character determination area, the initial scanned image is updated based on a preset image replacement strategy.

8. The method for digitalizing customer files according to claim 7, characterized in that: In response to the absence of any image character in the character determination area, updating the initial scanned image based on a preset image replacement strategy includes: In response to the absence of any image character in the character determination area, acquiring pixels of the character determination area, and determining the obtained regional pixel value corresponding to the character determination area as a reference pixel value; Acquire pixels of the noise area to obtain noise pixels constituting the noise area; Pixel replacement is performed on each noise pixel value corresponding to each noise pixel point based on the reference pixel value to obtain the updated initial scan image.

9. The method for digitalizing customer files according to claim 1, characterized in that: Performing content recognition on each current scanned image in the initial arrangement sequence, and determining whether there is an archive continuity relationship between each current scanned image in adjacent positions based on the recognition result, including: Performing paragraph recognition on each current scanned image in the initial arrangement sequence to obtain each character paragraph arranged vertically corresponding to the same current scanned image; Determining each current scan image as a primary scan image in sequence based on the initial arrangement sequence, and determining the current scan image located after the primary scan image as a secondary scan image; Determine the last character paragraph in the primary scanned image as the preceding connected paragraph, determine the first character paragraph in the secondary scanned image as the following connected paragraph, and obtain the last image character in the preceding connected paragraph and the first image character in the following connected paragraph; Retrieving a preset segmentation character group, wherein the preset segmentation character group includes a colon character, a period character, a semicolon character, and a comma character; The preset segmented character group is traversed to determine whether the last image character and the first image character are included in the preset segmented character group, and based on the determination result, it is determined whether there is an archival continuity relationship between the current scanned images at adjacent positions.

10. The method for digitalizing customer files according to claim 9, characterized in that: Judging whether there is an archive continuity relationship between the current scanned images at adjacent positions based on the determination result includes: In response to one of the last image character and the first image character being included in the preset segmented character group, determining a primary complete paragraph including the last image character and a secondary complete paragraph including the first image character in the primary scanned image and the secondary scanned image respectively; Performing semantic recognition on the primary complete paragraph and the secondary complete paragraph respectively, and determining whether there is an archival continuity relationship between the primary scanned image and the secondary scanned image based on the recognition result; In response to the last image character and the first image character not being included in the preset segmented character group, determining a primary complete paragraph including the last image character and a secondary complete paragraph including the first image character in the primary scanned image and the secondary scanned image respectively, and merging the primary complete paragraph and the secondary complete paragraph to obtain a merged complete paragraph; Semantic recognition is performed on the merged complete paragraph, and based on the recognition result, it is determined whether there is an archival continuity relationship between the primary scanned image and the secondary scanned image.

11. The method for digitalizing customer files according to claim 10, characterized in that: In response to the absence of an archival continuity relationship between any of the current scanned images at adjacent positions, abnormal marking is performed based on the initial arrangement sequence, and the obtained current arrangement sequence is sent to the user terminal for viewing, including: In response to the absence of an archival continuity relationship between any of the current scanned images at adjacent positions, a scan display layer is created, and scan display areas having the same number are established in the scan display layer based on the number of images corresponding to the current scanned images, wherein the scan display areas include a sequence display sub-area and an image display sub-area; Filling each current scan order into each order display sub-area located in the scan display layer and filling each current scan image into each image display sub-area located in the scan display layer based on the initial arrangement order; Establishing image enclosing frames respectively enclosing each current scanned image that does not have the archive connection relationship in the scan display layer, and forming a frame connection line between two image enclosing frames at adjacent positions, so as to complete the abnormal marking of the initial arrangement sequence; The obtained scanned display layer corresponding to the current arrangement sequence is sent to the user terminal for viewing.

12. A customer file digital processing system, characterized in that: include: The scanning module is configured to obtain a primary light collection value output by a primary light sensor disposed in a first collection sub-area corresponding to a preset collection area, and compare the primary light collection value with a retrieved preset light reference value; in response to the primary light collection value being less than the preset light reference value, trigger the acquisition of each secondary light collection value output by each secondary light sensor disposed in a second collection sub-area surrounding the first collection sub-area corresponding to the preset collection area, and compare each secondary light collection value with the preset light reference value; in response to each secondary light collection value being less than the preset light reference value, determine that the customer file is placed in the preset collection area, and trigger the scanning unit to sequentially perform scanning operations on each file page constituting the customer file; a recognition module configured to, in response to obtaining an initial scan image corresponding to any archive page based on the scanning operation, perform noise recognition on the initial scan image, and in response to determining a noise region existing in the initial scan image based on the recognition result, update the initial scan image to obtain a current scan image corresponding to a current scanning order; A sorting module is configured to, in response to receiving a scanning termination signal sent by the user terminal, trigger the scanning unit to stop the scanning operation, and sort all currently scanned images acquired based on the corresponding current scanning orders to obtain an initial arrangement sequence; a determination module configured to perform content recognition on each current scanned image in the initial arrangement sequence, and determine whether there is an archival continuity relationship between each current scanned image in adjacent positions based on the recognition result; The marking module is configured to perform an abnormal marking based on the initial arrangement sequence in response to the absence of an archival continuity relationship between any of the current scanned images in adjacent positions, and send the obtained current arrangement sequence to the user terminal for viewing.

Images