An OCR-based automatic recognition and uploading system and device for freight weighbridge slips

Abstract

This invention relates to the field of recognition technology, specifically to an OCR-based automatic recognition and uploading system and device for freight weighbridge slips. The system includes a detection base, a support frame, a fixed image acquisition unit, a mobile phone clamping assembly, and a weighbridge slip backing plate. The support frame is fixedly mounted on the detection base, and the fixed image acquisition unit and mobile phone clamping assembly are mounted on the support frame. The weighbridge slip backing plate is detachably placed on the detection base via a positioning mechanism. A mechanism to prevent misoperation ensures the driver's correct operation when scanning the weighbridge slip. The accuracy of weighbridge slip recognition is improved by using a weighbridge slip backing plate designed according to the corresponding type of weighbridge slip.

Description

Technical Field

[0001] This invention relates to the field of identification technology, specifically to an OCR-based automatic identification and uploading system and device for freight weighbridge slips. Background Technology

[0002] The identification of freight weighbridge slips is a crucial step in the entire transportation process and a key basis for later settlement. Drivers must carefully check their weighbridge slips after each day's transport and record the net weight of the furs transported. Due to the unique environment of the transportation industry, variations in time and environment, and differences in drivers' educational levels, the error rate in entering freight weights is relatively high, significantly increasing the workload for later verification. Automated weighbridge slip identification would not only reduce the workload for drivers and document reviewers but also represent a significant leap towards intelligent management of the entire process. It plays an indispensable role in subsequent automated document review and other processes.

[0003] Currently, most weighbridge slips are entered manually or through simple image capture technology. The drawback of this approach is that it can only recognize a limited number of different slip formats. Different slip formats require different processing logic, making it unreliable. What is needed is a technological solution that overcomes the limitations of different weighbridge slip formats, providing a universal processing method that simplifies driver procedures, prevents driver errors, and improves data accuracy.

[0004] Current image recognition technologies on the market all use general APIs to capture image data and then process the image data based on keywords for recognition. This method can only recognize a limited number of templates and has low usability. Therefore, it is necessary to combine image recognition technology with the rules of the weighbridge slip to achieve a breakthrough. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an OCR-based automatic identification and uploading system and device for freight weighbridge slips. By setting up a mechanism to prevent misoperation to ensure the correctness of the driver's operation when scanning the weighbridge slip, and by setting a weighbridge slip backing plate designed according to the corresponding type of weighbridge slip, the accuracy of weighbridge slip identification is improved.

[0006] Based on the characteristics of weighbridge slips—the rule that tare weight + net weight = gross weight—this invention designs an algorithm to extract data that conforms to the rule, and then processes it according to the corresponding rules to obtain the corresponding tare weight, net weight, and gross weight. This achieves a universal function that works with different weighbridge slip templates, improving the limitations of weighbridge slip recognition on templates and making it more universal and easier to use.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] An OCR-based automatic recognition and uploading device for freight weighbridge slips includes a detection base, a support frame, a fixed image acquisition unit, a mobile phone clamping component, and a weighbridge slip liner. The support frame is fixedly mounted on the detection base, and the fixed image acquisition unit and the mobile phone clamping component are mounted on the support frame. The weighbridge slip liner is detachably placed on the detection base via a positioning mechanism.

[0009] The fixed image capturing unit and the mobile phone clamping assembly are set above the weighbridge sheet liner through a support frame. The weighbridge sheet to be identified is placed on the weighbridge sheet liner, and the image of the weighbridge sheet to be identified is acquired by the mobile phone device on the fixed image capturing unit / mobile phone clamping assembly.

[0010] The mobile phone clamping component can adjust its relative position to the weighbridge liner in both the horizontal and vertical directions, and can also adjust the relative distance between the two sides.

[0011] The fixed image acquisition unit can be rotated to a working position and a folded position. When the fixed image acquisition unit is in the folded position, the mobile phone device is clamped on the mobile phone clamping component, the mobile phone device is used to acquire the weighbridge image, and the weighbridge image is processed by the third-party software therein.

[0012] When the fixed image acquisition unit is in the working position, the weighbridge image is acquired through the fixed image acquisition unit, and the fixed image acquisition unit is fixed to the support frame through the first locking structure. At the same time, the mobile phone clamping component is fixed to the fixed image acquisition unit through the second locking structure.

[0013] Furthermore, when the fixed image capturing part rotates from the folded position to the working position, it is locked and fixed to the mobile phone clamping component from below by the second locking structure;

[0014] When the fixed image capturing unit is in the working position, the movement of the mobile phone clamping component can be restricted by the second locking part to prevent operator misoperation.

[0015] The extension direction of the fixed image capturing part partially coincides with the clamping position of the mobile phone device on the mobile phone clamping assembly, so that the fixed image capturing part can block the shooting path of the mobile phone device when it is in the working position, further preventing operator misoperation.

[0016] Furthermore, the fixed image capturing unit includes a rotating bracket, a hinged base, a fixed camera assembly, a fill light, a first pin, and a second pin;

[0017] The rotating bracket is rotatably mounted in the middle of the second crossbeam of the support frame via a hinged seat. An adjacent fixed camera assembly and a supplementary light are provided on the rotating bracket facing the weighing slip liner.

[0018] A first pin is provided at one end of the rotating bracket away from the hinge seat, and a first locking part is provided in the middle of the first crossbeam of the support frame. When in the working position, the first pin engages with the first locking part to lock the rotating bracket.

[0019] A second pin is provided in the middle of the rotating bracket, and a second locking part is provided below the mobile phone clamping assembly. When in the working position, the second pin engages with the second locking part to lock the mobile phone clamping assembly.

[0020] Furthermore, the mobile phone clamping assembly also includes a movable crossbar, a first sliding support column, a second sliding support column, a translational bottom rail, a translational top rail, a sliding block, a fixed clamping part, and a movable clamping part;

[0021] Two translational bottom rails are arranged on both sides of the detection base. The first sliding support column and the second sliding support column are slidably supported on the translational bottom rails respectively. The two ends of the moving crossbar are slidably engaged with the first sliding support column and the second sliding support column respectively through sliding structures. The moving crossbar can slide vertically on the first sliding support column and the second sliding support column.

[0022] A second locking part is provided in the middle of the lower end face of the moving crossbar, and a transverse top rail is provided on the upper end face of the moving crossbar. A sliding block is connected to a fixed clamping part on the transverse top rail, and a movable clamping part is connected to the fixed clamping part through an operating part.

[0023] The mobile phone device is held in place by the fixed clamping part and the movable clamping part.

[0024] Furthermore, the detection base includes a base, a diffuser cover, a first positioning groove, a second positioning groove, an LED cold light source, and a reflector;

[0025] The weighing slip liner includes a plate body, limiting protrusions, positioning blocks, information area recognition texture, and bottom background texture;

[0026] The upper surface of the plate is provided with a limiting protrusion to achieve positioning of the weighbridge slip to be identified. The lower surface of the plate is provided with a positioning block. The upper surface of the plate is also provided with an information area recognition texture and a bottom background texture.

[0027] The base is provided with multiple ring-shaped reflectors, and LED cold light sources are fixed on the reflectors. A diffuser cover is provided on the LED cold light source. A first positioning groove and a second positioning groove are provided at different positions on the diffuser cover. The first positioning groove and the second positioning groove correspond to the positioning blocks of weighing slip liners of different sizes.

[0028] Furthermore, the weighing slip backing is made of transparent material. After the LED cold light source is reflected by the reflector and scattered by the diffuser cover, the information area recognition texture and the bottom background texture on the weighing slip backing can be acquired by the fixed image capturing unit / mobile device.

[0029] Each weighing slip backing plate corresponds to a type of weighing slip to be identified, and corresponds to different forms of information area identification texture and bottom background texture, so as to realize the location of important information on the weighing slip to be identified.

[0030] Furthermore, when the weighbridge slip to be identified is damaged, the information area recognition texture / bottom background texture on the weighbridge slip backing can be obtained by the fixed image acquisition unit / mobile device along with the image information of the weighbridge slip through the damaged location of the weighbridge slip to be identified, thereby determining the damaged location of the weighbridge slip to be identified and eliminating the impact of the damage on image recognition during subsequent image processing.

[0031] An OCR-based automatic recognition and uploading system for freight weighbridge slips, employing the aforementioned automatic recognition and uploading device, includes an image recognition module. The image recognition module is connected to a mobile device with a fixed image capture unit and a mobile phone clamping component. The image recognition module first acquires a textured table image on the weighbridge slip backing, determines and generates the shape, position, and arrangement information of specific cell groups, and simultaneously generates a bottom texture image based on the bottom background texture on the weighbridge slip backing.

[0032] The weighbridge slip to be identified is placed on the weighbridge slip backing plate, and the weighbridge slip is positioned by the limiting protrusion. Then the image recognition module acquires the weighbridge slip image information.

[0033] The fixed image capturing unit transmits the acquired weighbridge image information to the image recognition module, and the mobile device of the mobile phone clamping component sends the acquired weighbridge image information to the image recognition module after processing by a third-party data terminal.

[0034] The image recognition module uses an initial hypertext markup language formed based on the specific cell group to describe the table structure of the bill image and recognizes the strings in the bill image; it determines whether a bottom texture image appears in the bill image, and corrects the strings based on the boundary pixels of the bottom texture image to form a new corrected string;

[0035] Determine whether the positions of the detection box of the cell and the detection box of the correction string are closest and whether the intersection-union ratio of the detection boxes is the largest. Then, insert the identified correction string into the initial hypertext markup language corresponding to the cell and generate the structured dataset of the bill of lading to be identified based on the hypertext markup language.

[0036] Furthermore, the process of correcting the string based on the boundary pixels of the bottom texture image to form a new corrected string includes: analyzing and determining each pixel in the image of the string, determining each pixel in the bottom texture image at the position corresponding to the string, removing all pixels in the bottom texture image from the image of the string, and generating the corrected string based on the removed string image.

[0037] Furthermore, the structured dataset of the weighbridge slip includes tare weight data, net weight data, and gross weight data. Based on the rule that the sum of tare weight data and net weight data equals the gross weight data, the structured dataset of the identified weighbridge slip is verified. If the verification result is incorrect, the weighbridge slip to be identified is re-identified.

[0038] Compared with existing technologies, the present invention provides an OCR-based automatic recognition and uploading system and device for freight weighbridge slips, which has the following advantages:

[0039] 1. The automatic weighbridge receipt recognition device of the present invention has a fixed image acquisition unit and a mobile phone clamping component. It can acquire weighbridge receipt images through the fixed image acquisition unit and send them to the built-in image processing system. It can also be adapted to other weighbridge receipt recognition systems installed on the driver's mobile phone, thus adapting to various types of freight trucks and different settlement systems.

[0040] 2. The automatic weighbridge slip recognition device of the present invention is equipped with a mechanism to prevent misoperation, which can restrict the operation of the mobile phone clamping component and the mobile phone device thereon when the fixed image capturing unit is working, thereby reducing the difficulty of operation and recognition for the driver.

[0041] 3. This invention designs a corresponding weighbridge backing for a certain type of weighbridge slip. The weighbridge backing has specific textures on the important information positions (such as tare weight, net weight, gross weight, and product name), and a background texture is also set on the bottom background. This can more effectively identify the table rules in this type of weighbridge slip and improve the recognition accuracy.

[0042] 4. The texture on the weighbridge slip backing of the present invention can effectively mark the location of damage when the weighbridge slip is damaged. The damaged slip will be captured by a camera at the damaged area, and the image of the damaged location can be corrected by a specially designed image processing method, thereby further improving the recognition accuracy of OCR.

[0043] 5. Through our analysis of the characteristics of different weighbridge slips, we have implemented a universal processing logic capable of recognizing weighbridge slips of various formats. This breaks down the barriers of fixed templates in weighbridge slip recognition. It can be effectively applied to our transportation operations, improving efficiency, reducing operational steps for staff, and resulting in cleaner data.

[0044] 6. Based on the characteristics of weighbridge slips—tare weight + net weight = gross weight—an algorithm was designed to extract data that conforms to this rule. This data is then processed according to the corresponding rules to obtain the tare weight, net weight, and gross weight. This algorithm enables universal compatibility with different weighbridge slip templates, improving the limitations of template-based weighbridge slip recognition and making it more versatile and easier to use. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of two types of weighbridge slips to be identified according to the present invention;

[0046] Figure 2 This is a schematic diagram of the automatic weighbridge slip recognition device of the present invention;

[0047] Figure 3 This is a schematic diagram of the fixed image capturing unit of the present invention;

[0048] Figure 4 This is a schematic diagram of the structure of the mobile phone clamping component of the present invention;

[0049] Figure 5 This is a schematic diagram of the weighbridge liner of the present invention;

[0050] Figure 6 A schematic diagram showing a weighing slip liner of another size assembled for this invention;

[0051] Figure 7 This is a schematic diagram of the detection base of the present invention;

[0052] Figure 8 This is a schematic diagram of the internal structure of the detection base of the present invention;

[0053] In the picture:

[0054] Detection base 1, base 11, diffuser cover 12, first positioning groove 121, second positioning groove 122, LED cold light source 13, reflector 14;

[0055] Support frame 2, first crossbeam 21, second crossbeam 22, first locking part 23;

[0056] Fixed image capturing unit 3, rotating bracket 31, hinge seat 32, fixed camera assembly 33, supplementary light 34, first pin 35, second pin 36;

[0057] Mobile phone clamping assembly 4, moving crossbar 41, second locking part 42, first sliding support column 43, second sliding support column 44, translation bottom rail 45, translation top rail 46, sliding block 47, fixed clamping part 48, movable clamping part 49;

[0058] 5. Weighing slip backing plate; 51. Plate body; 52. Limiting protrusion; 53. Positioning block; 54. Information area recognition texture; 55. Bottom background texture; Detailed Implementation

[0059] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0060] The following is based on the appendix Figure 1-8 This invention provides a detailed description of an OCR-based automatic recognition and uploading device for freight weighbridge slips. The device includes a detection base 1, a support frame 2, a fixed image acquisition unit 3, a mobile phone clamping assembly 4, and a weighbridge slip backing plate 5. The support frame 2 is fixedly mounted on the detection base 1, and the fixed image acquisition unit 3 and the mobile phone clamping assembly 4 are mounted on the support frame 2. The weighbridge slip backing plate 5 is detachably placed on the detection base 1 via a positioning mechanism. The fixed image acquisition unit 3 and the mobile phone clamping assembly 4 are positioned above the weighbridge slip backing plate 5 via the support frame 2. The weighbridge slip to be recognized is placed on the weighbridge slip backing plate 5, and the mobile phone device on the fixed image acquisition unit 3 / mobile phone clamping assembly 4 acquires the weighbridge slip information. The image is a single piece of paper; the mobile phone clamping component 4 can adjust the relative horizontal and vertical position of the distance from the weighbridge sheet 5, and can also adjust the relative distance; the fixed image capturing unit 3 can rotate to the working position and the folded position. When the fixed image capturing unit 3 is in the folded position, the mobile phone device is clamped on the mobile phone clamping component 4 to acquire the weighbridge sheet image, and the weighbridge sheet image is processed by the third-party software therein; when the fixed image capturing unit 3 is in the working position, the weighbridge sheet image is acquired through the fixed image capturing unit 3, and the fixed image capturing unit 3 is fixed to the support frame 2 by the first locking structure, while the mobile phone clamping component 4 is fixed to the fixed image capturing unit 3 by the second locking structure.

[0061] The fixed image acquisition unit 3 and the mobile phone clamping component 4 in the above-mentioned automatic weighbridge recognition device can not only acquire weighbridge images through the fixed image acquisition unit 3 and send them to the built-in image processing system, but also be adapted to other weighbridge recognition systems installed on the driver's mobile phone, thus adapting to various types of freight trucks and different settlement systems.

[0062] Furthermore, when the fixed image capturing part 3 rotates from the folded position to the working position, it is locked and fixed to the mobile phone clamping assembly 4 from below by the second locking structure; when the fixed image capturing part 3 is in the working position, it can restrict the movement of the mobile phone clamping assembly 4 by the second locking part to prevent operator misoperation; the extension direction of the fixed image capturing part 3 partially coincides with the clamping position of the mobile phone device on the mobile phone clamping assembly 4, so that the fixed image capturing part 3 can block the shooting path of the mobile phone device when it is in the working position, further preventing operator misoperation.

[0063] The aforementioned mechanism for preventing misoperation can restrict the operation of the mobile phone clamping assembly 4 and the mobile phone device thereon when the fixed image capturing unit 3 is working, thereby reducing the difficulty of operation and recognition for the driver.

[0064] Furthermore, the fixed image capturing unit 3 includes a rotating bracket 31, a hinge seat 32, a fixed camera assembly 33, a fill light 34, a first pin 35, and a second pin 36. The rotating bracket 31 is rotatably mounted in the middle of the second crossbeam 22 of the support frame 2 via the hinge seat 32. The fixed camera assembly 33 and the fill light 34 are arranged adjacent to each other on the rotating bracket 31 facing the weighing slip liner 5. The first pin 35 is provided at the end of the rotating bracket 31 away from the hinge seat 32. A first locking part 23 is provided in the middle of the first crossbeam 21 of the support frame 2. When in the working position, the first pin 35 engages with the first locking part 23 to lock the rotating bracket 31. The second pin 36 is provided in the middle of the rotating bracket 31. A second locking part 42 is provided below the mobile phone clamping assembly 4. When in the working position, the second pin 36 engages with the second locking part 42 to lock the mobile phone clamping assembly 4.

[0065] Furthermore, the mobile phone clamping assembly 4 also includes a movable crossbar 41, a first sliding support column 43, a second sliding support column 44, a translational bottom rail 45, a translational top rail 46, a sliding block 47, a fixed clamping part 48, and a movable clamping part 49; the two translational bottom rails 45 are disposed on both sides of the detection base 1, and the first sliding support column 43 and the second sliding support column 44 are slidably supported on the translational bottom rails 45 respectively, and the two ends of the movable crossbar 41 are respectively connected to the first sliding support column 43 and the second sliding support column 44 through sliding structures. The column 44 is slidably engaged, and the movable crossbar 41 can slide vertically on the first sliding support column 43 and the second sliding support column 44; a second locking part 42 is provided in the middle of the lower end face of the movable crossbar 41, and a transverse top rail 46 is provided on the upper end face of the movable crossbar 41. The transverse top rail 46 is connected to the fixed clamping part 48 through a sliding block 47, and the fixed clamping part 48 is connected to a movable movable clamping part 49 through an operating part; the mobile phone device is clamped and fixed by the fixed clamping part 48 and the movable clamping part 49.

[0066] Furthermore, the detection base 1 includes a base 11, a diffuser cover 12, a first positioning groove 121, a second positioning groove 122, an LED cold light source 13, and a reflector 14; the weighbridge slip backing plate 5 includes a plate body 51, a limiting protrusion 52, a positioning block 53, an information area recognition texture 54, and a bottom background texture 55; the upper end surface of the plate body 51 is provided with a limiting protrusion 52 to achieve positioning of the weighbridge slip to be identified, the lower end surface of the plate body 51 is provided with a positioning block 53, and the upper end of the plate body 51... The surface is also provided with an information area identification texture 54 and a bottom background texture 55; wherein the base 11 is provided with a plurality of ring-shaped reflectors 14, an LED cold light source 13 is fixed on the reflectors 14, a diffuser cover 12 is provided on the LED cold light source 13, and a first positioning groove 121 and a second positioning groove 122 are provided at different positions on the diffuser cover 12, and the first positioning groove 121 and the second positioning groove 122 respectively correspond to the positioning blocks 53 of the weighing slip liner 5 of different sizes.

[0067] For each type of weighbridge slip, a corresponding weighbridge slip backing plate 5 is designed. The weighbridge slip backing plate 5 has specific textures set on the important information positions (such as tare weight, net weight, gross weight, goods name, etc.), and a background texture is also set on the bottom background. This can more effectively identify the table rules in this type of weighbridge slip and improve the recognition accuracy.

[0068] Furthermore, the weighbridge slip backing plate 5 is made of transparent material. After the LED cold light source 13 is reflected by the reflector 14 and scattered by the diffuser cover 12, the information area recognition texture 54 and the bottom background texture 55 on the weighbridge slip backing plate 5 can be acquired by the fixed image capturing unit 3 / mobile device. Each weighbridge slip backing plate 5 corresponds to a type of weighbridge slip to be identified and corresponds to different forms of information area recognition texture 54 and bottom background texture 55, so as to realize the positioning of important information on the weighbridge slip to be identified.

[0069] Furthermore, when the weighbridge slip to be identified is damaged, the information area recognition texture 54 / bottom background texture 55 on the weighbridge slip backing plate 5 can be obtained by the fixed image acquisition unit 3 / mobile device along with the image information of the weighbridge slip through the damaged location of the weighbridge slip to be identified, thereby determining the damaged location of the weighbridge slip to be identified and eliminating the impact of the damage on image recognition during subsequent image processing.

[0070] The texture on the weighbridge slip backing plate 5 can effectively mark the location of the damage when the weighbridge slip is damaged. The damaged slip will be captured by the camera at the damaged area, and the image of the damaged location can be corrected through a specially designed image processing method, which further improves the recognition accuracy of OCR.

[0071] An OCR-based automatic recognition and uploading system for freight weighbridge slips, employing the aforementioned automatic recognition and uploading device, includes an image recognition module. The image recognition module is connected to a mobile device with a fixed image capturing unit 3 and a mobile phone clamping component 4. The image recognition module first acquires the textured table image on the weighbridge slip backing plate 5, determines and generates the shape, position, and arrangement information of specific cell groups, and simultaneously generates a bottom texture image based on the bottom background texture 55 on the weighbridge slip backing plate 5.

[0072] The weighbridge slip to be identified is placed on the weighbridge slip backing plate 5, and the weighbridge slip is positioned by the limiting protrusion 52. Then the image recognition module obtains the image information of the weighbridge slip.

[0073] The fixed image capturing unit 3 transmits the acquired weighbridge image information to the image recognition module, and the mobile phone device of the mobile phone clamping component 4 sends the acquired weighbridge image information to the image recognition module after processing by a third-party data terminal.

[0074] The image recognition module uses an initial hypertext markup language formed based on the specific cell group to describe the table structure of the bill image and recognizes the strings in the bill image; it determines whether a bottom texture image appears in the bill image, and corrects the strings based on the boundary pixels of the bottom texture image to form a new corrected string;

[0075] Determine whether the positions of the detection box of the cell and the detection box of the correction string are closest and whether the intersection-union ratio of the detection boxes is the largest. Then, insert the identified correction string into the initial hypertext markup language corresponding to the cell and generate the structured dataset of the bill of lading to be identified based on the hypertext markup language.

[0076] Furthermore, the process of correcting the string based on the boundary pixels of the bottom texture image to form a new corrected string includes: analyzing and determining each pixel in the image of the string, determining each pixel in the bottom texture image at the position corresponding to the string, removing all pixels in the bottom texture image from the image of the string, and generating the corrected string based on the removed string image.

[0077] Furthermore, the structured dataset of the weighbridge slip includes tare weight data, net weight data, and gross weight data. Based on the rule that the sum of tare weight data and net weight data equals the gross weight data, the structured dataset of the identified weighbridge slip is verified. If the verification result is incorrect, the weighbridge slip to be identified is re-identified.

[0078] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An OCR-based automatic recognition and uploading device for freight weighbridge slips, comprising a detection base (1), a support frame (2), a fixed image acquisition unit (3), a mobile phone clamping assembly (4), and a weighbridge slip backing plate (5), characterized in that: A support frame (2) is fixedly installed on the detection base (1). A fixed image acquisition unit (3) and a mobile phone clamping assembly (4) are installed on the support frame (2). The weighing slip liner (5) is detachably placed on the detection base (1) through a positioning mechanism. The fixed image capturing unit (3) and the mobile phone clamping assembly (4) are set above the weighing slip liner (5) through the support frame (2). The weighing slip to be identified is placed on the weighing slip liner (5), and the image of the weighing slip to be identified is obtained by the mobile phone device on the fixed image capturing unit (3) / mobile phone clamping assembly (4). The mobile phone clamping component (4) can adjust the relative position of the weighing slip backing plate (5) in the horizontal and vertical directions, and can also adjust the relative distance; The fixed image acquisition unit (3) can be rotated to the working position and the folded position. When the fixed image acquisition unit (3) is in the folded position, the mobile phone device is clamped on the mobile phone clamping component (4), the mobile phone device is used to acquire the weighing slip image, and the weighing slip image is processed by the third-party software therein. When the fixed image acquisition unit (3) is in the working position, the weighing slip image is acquired through the fixed image acquisition unit (3), and the fixed image acquisition unit (3) is fixed to the support frame (2) through the first locking structure, while the mobile phone clamping assembly (4) is fixed to the fixed image acquisition unit (3) through the second locking structure. When the fixed image capturing part (3) rotates from the folded position to the working position, it is locked and fixed to the mobile phone clamping assembly (4) from below by the second locking structure; When the fixed image capturing unit (3) is in the working position, it can restrict the movement of the mobile phone clamping assembly (4) by means of the second locking part to prevent operator misoperation; The extension direction of the fixed image capturing part (3) partially coincides with the clamping position of the mobile phone device on the mobile phone clamping assembly (4), so that the fixed image capturing part (3) can block the shooting path of the mobile phone device when it is in the working position, further preventing operator misoperation. The fixed image capturing unit (3) includes a rotating bracket (31), a hinge base (32), a fixed camera assembly (33), a fill light (34), a first pin (35), and a second pin (36); The rotating bracket (31) is rotatably mounted in the middle of the second crossbeam (22) of the support frame (2) via a hinge seat (32). An adjacent fixed camera assembly (33) and a supplementary light (34) are provided on the rotating bracket (31) facing the weighing slip liner (5). The rotating bracket (31) is provided with a first pin (35) at one end away from the hinge seat (32), and the first crossbeam (21) of the support frame (2) is provided with a first locking part (23) in the middle. When in the working position, the first pin (35) and the first locking part (23) are engaged to lock the rotating bracket (31). The rotating bracket (31) is provided with a second pin (36) in the middle, and the mobile phone clamping assembly (4) is provided with a second locking part (42) below. When in the working position, the second pin (36) and the second locking part (42) engage to lock the mobile phone clamping assembly (4).

2. The OCR-based automatic recognition and uploading device for freight weighbridge slips according to claim 1, characterized in that: The mobile phone clamping assembly (4) also includes a moving crossbar (41), a first sliding support column (43), a second sliding support column (44), a translational bottom rail (45), a translational top rail (46), a sliding block (47), a fixed clamping part (48), and a movable clamping part (49). Two translational bottom rails (45) are disposed on both sides of the detection base (1). The translational bottom rails (45) respectively support the first sliding support column (43) and the second sliding support column (44). The two ends of the moving crossbar (41) are slidably engaged with the first sliding support column (43) and the second sliding support column (44) through a sliding structure. The moving crossbar (41) can slide vertically on the first sliding support column (43) and the second sliding support column (44). A second locking part (42) is provided in the middle of the lower end face of the moving crossbar (41), and a transverse top rail (46) is provided on the upper end face of the moving crossbar (41). The transverse top rail (46) is connected to the fixed clamping part (48) by a sliding block (47), and the fixed clamping part (48) is connected to the movable clamping part (49) by an operating part. The mobile phone device is held in place by the fixed clamping part (48) and the movable clamping part (49).

3. The OCR-based automatic recognition and uploading device for freight weighbridge slips according to claim 2, characterized in that: The detection base (1) includes a base (11), a diffuser cover (12), a first positioning groove (121), a second positioning groove (122), an LED cold light source (13), and a reflector (14). The weighing slip liner (5) includes a plate body (51), a limiting protrusion (52), a positioning block (53), an information area recognition texture (54), and a bottom background texture (55). The upper end face of the plate (51) is provided with a limiting protrusion (52) to realize the positioning of the weighbridge slip to be identified. The lower end face of the plate (51) is provided with a positioning block (53). The upper end face of the plate (51) is also provided with an information area identification texture (54) and a bottom background texture (55). The base (11) is provided with a plurality of ring-shaped reflectors (14), and LED cold light sources (13) are fixed on the reflectors (14). A diffuser cover (12) is provided on the LED cold light sources (13). A first positioning groove (121) and a second positioning groove (122) are provided at different positions on the diffuser cover (12). The first positioning groove (121) and the second positioning groove (122) correspond to the positioning blocks (53) of the weighing slip liner (5) of different sizes.

4. The OCR-based automatic recognition and uploading device for freight weighbridge slips according to claim 3, characterized in that: The weighing slip backing plate (5) is made of transparent material. After the LED cold light source (13) is reflected by the reflector (14) and scattered by the diffuser cover plate (12), the information area identification texture (54) and the bottom background texture (55) on the weighing slip backing plate (5) can be acquired by the fixed image capturing unit (3) / mobile device. Each of the weighing slip backing plates (5) corresponds to a type of weighing slip to be identified, and corresponds to different forms of information area identification texture (54) and bottom background texture (55) to realize the positioning of important information on the weighing slip to be identified.

5. The OCR-based automatic recognition and uploading device for freight weighbridge slips according to claim 4, characterized in that: When the weighbridge slip to be identified is damaged, the information area identification texture (54) / bottom background texture (55) on the weighbridge slip backing plate (5) can be obtained by the fixed image acquisition unit (3) / mobile device along with the image information of the weighbridge slip through the damaged location of the weighbridge slip to be identified, thereby determining the damaged location of the weighbridge slip to be identified and eliminating the impact of the damage on image recognition during subsequent image processing.

6. An OCR-based automatic identification and uploading system for freight weighbridge slips, employing the automatic identification and uploading device described in any one of claims 1-5, characterized in that: The image recognition module is connected to a mobile phone device with a fixed image capturing unit (3) and a mobile phone clamping component (4). The image recognition module first acquires the texture table image on the weighing slip backing board (5), determines and generates the shape, position and arrangement information of a specific cell group, and generates a bottom texture image based on the bottom background texture (55) on the weighing slip backing board (5). The weighbridge slip to be identified is placed on the weighbridge slip backing plate (5), and the weighbridge slip is positioned by the limiting protrusion (52). Then the image recognition module obtains the weighbridge slip image information. The fixed image capturing unit (3) transmits the acquired weighbridge image information to the image recognition module, and the mobile phone device of the mobile phone clamping component (4) sends the acquired weighbridge image information to the image recognition module after processing by a third-party data terminal. The image recognition module uses an initial hypertext markup language formed based on the specific cell group to describe the table structure of the bill image and recognizes the strings in the bill image; Determine whether a bottom texture image appears in the weighing slip image, and correct the string based on the boundary pixels of the bottom texture image to form a new corrected string; Determine whether the positions of the detection box of the specific cell and the detection box of the correction string are closest and whether the intersection-union ratio of the detection boxes is the largest. Then, insert the identified correction string into the initial hypertext markup language corresponding to the specific cell, and generate the structured dataset of the bill of lading to be identified based on the hypertext markup language.

7. The automatic identification and uploading system according to claim 6, characterized in that: The process of correcting the string based on the boundary pixels of the bottom texture image to form a new corrected string includes: analyzing and determining each pixel in the image of the string; determining each pixel in the bottom texture image at the position corresponding to the string; removing all pixels in the bottom texture image from the image of the string; and generating the corrected string based on the removed string image.

8. The automatic identification and uploading system according to claim 7, characterized in that: The structured dataset of the weighbridge slips includes tare weight data, net weight data, and gross weight data. The structured dataset of the identified weighbridge slips is verified according to the rule that the sum of the tare weight data and the net weight data equals the gross weight data. If the verification result is incorrect, the weighbridge slips to be identified are re-identified.

Images