A self-service book borrowing machine with a touch transparent screen

Abstract

The application belongs to the technical field of library automation equipment, and particularly relates to a touch transparent screen self-service book borrowing machine for books, which comprises a cabinet body, a touch transparent screen and a book returning port, a multi-dimensional fusion intelligent detection book integrity detection module is integrated in the book returning port, the module automatically turns the pages of the returned book through a flexible page turning mechanism, and uses an infrared transmission scanning, thickness / pressure distribution sensing, three-dimensional topography scanning and spectral reflectance analysis submodule arranged around to synchronously collect multi-dimensional physical characteristic data of the pages, an AI analysis model in a central processing module carries out fusion comparison analysis on real-time data and pre-stored standard characteristic data to generate an accurate inner page integrity diagnosis report. The application realizes instant, non-destructive and automatic detection of problems such as missing pages, damage and writing on the inner pages of the returned book, and greatly improves the asset management efficiency and intelligent level of the library.

Description

Technical Field

[0001] This invention relates to the field of library automation equipment technology, and in particular to a self-service book borrowing machine with a transparent touch screen. Background Technology

[0002] Self-service book lending machines have become an important part of modern libraries. By integrating RFID (Radio Frequency Identification) or barcode scanning technology, they allow readers to borrow and return books independently, significantly improving service efficiency and reducing labor costs. Existing self-service book lending machines typically include: a touchscreen for user interaction, a book identification module (such as an RFID reader or barcode scanner), a book delivery mechanism, and a control unit.

[0003] However, existing technology has a significant drawback: when readers return books, the equipment can only identify the book's identity information (i.e., "which book it is"), but cannot automatically and efficiently detect the book's internal physical condition, especially the integrity of the pages. Problems such as missing pages (e.g., torn), damaged pages (e.g., torn corners, cuts, graffiti, water stains, severe creases) currently rely entirely on librarians to manually check and inspect the books during subsequent organization. This method is inefficient, prone to oversights, and when problems are discovered, it is often difficult to trace the specific person responsible, leading to difficulties in determining book assets and liability.

[0004] Therefore, there is an urgent need for a self-service book lending machine that can automatically, without damage, and accurately detect the integrity of the book's inner pages upon return, in order to fill the gap in existing technology. Summary of the Invention

[0005] Based on the technical problems existing in the prior art, this invention proposes a self-service book lending machine with a transparent touch screen.

[0006] The present invention proposes a self-service book lending machine with a transparent touch screen, including a cabinet, a transparent touch screen installed on the cabinet, a control unit, and a book storage and retrieval mechanism installed inside the cabinet. The book storage and retrieval mechanism includes at least one borrowing slot and one returning slot. Inside the returning slot, a multi-dimensional fusion intelligent detection module for detecting the integrity of books is integrated along the book conveying path. The detection module includes: A flexible page-turning mechanism is used to automatically turn the pages during testing; A multi-sensor fusion detection array is arranged around the flexible page-turning mechanism to collect multi-dimensional physical feature data of the pages being turned; The central processing module is connected to the control unit and has a pre-installed AI analysis model. It is used to receive real-time data from the multi-sensor fusion detection array, compare and analyze the data with pre-stored standard feature data, and generate a book page integrity diagnostic report. The touch-sensitive transparent screen is configured to dynamically display the detection process and show the diagnostic report.

[0007] Preferably, the multi-sensor fusion detection array includes at least an infrared transmission scanning submodule, a thickness / pressure distribution sensing submodule, a three-dimensional topography scanning submodule, and a spectral reflectance analysis submodule.

[0008] Preferably, the infrared transmission scanning submodule includes an infrared linear light source and a high-resolution linear array camera, which are respectively set on both sides of the page transmission path, and are used to detect missing pages and large-area tears by analyzing the difference in infrared light transmission.

[0009] Preferably, the thickness / pressure distribution sensing submodule includes a flexible thin-film pressure distribution sensor laid on the bearing surface of the book return channel, used to acquire the pressure distribution map of the book pages to detect local defects and foreign objects.

[0010] Preferably, the three-dimensional shape scanning submodule includes a miniature structured light projector and a camera, used to reconstruct the three-dimensional shape of the page surface using structured light technology to detect dents and creases.

[0011] Preferably, the spectral reflectance analysis submodule includes a multispectral LED light source and a miniature spectrometer, used to detect handwriting and stains by analyzing the reflectance spectral characteristics of the page surface.

[0012] Preferably, the AI ​​analysis model in the central processing module is a multi-task fusion neural network, used to perform feature-level fusion and correlation analysis on data from the multi-sensor fusion detection array.

[0013] Preferably, the touch-sensitive transparent screen uses augmented reality technology to overlay the virtual book model with the real-time detection process animation, and makes a visual annotation at the corresponding position of the virtual model when damage is detected.

[0014] Preferably, the cabinet is further provided with a positioning and pushing device controlled by the control unit. The positioning and pushing device is driven by a linear mechanism to move the push plate. The positioning and pushing device triggers the positioning function at the detection station according to the instruction, or triggers the sorting function together with the conveyor roller group at the feeding station.

[0015] Preferably, it also includes an air blowing device installed on one side of the book return channel for use in conjunction with the flexible page turning mechanism. The air blowing device uses an inclined nozzle to stabilize the turned pages at an inclined angle.

[0016] Compared with the prior art, the present invention provides a self-service book lending machine with a transparent touch screen, which has the following advantages: It enables instant physical examination of returned books, upgrading the function of self-service borrowing and returning machines from simple identity recognition to intelligent diagnosis of the physical condition of books, fundamentally solving the problem of difficulty in timely detection of damaged inner pages.

[0017] The detection is comprehensive and accurate. Through multi-sensor (infrared, pressure, 3D, spectral) fusion technology, it cross-verifies from multiple dimensions such as light transmittance, thickness, morphology, and material composition, which greatly improves the accuracy of identifying various types of damage (missing pages, torn corners, writing, stains, creases) and the ability to resist interference.

[0018] The process is transparent and trustworthy, utilizing a touch-sensitive transparent screen for augmented reality (AR) visualization, making the testing process visible to users, enhancing the credibility of the equipment, and reducing potential disputes.

[0019] It is highly automated and efficient, integrating page turning, detection, judgment and sorting into one, with the entire process automated. All operations are completed in a short time (usually within 1 minute) when the user returns the book, without affecting the user experience. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the overall structure of the present invention; Figure 2 This is a side view of the structure of the present invention; Figure 3 This is a schematic diagram of the internal structure of the book return channel of the present invention; Figure 4 This is a schematic diagram of the state when the positioning and pushing device of the present invention performs the positioning function; Figure 5 This is a schematic diagram showing the state of the pages of the book as they are just turned up by the friction rollers. Figure 6 This is a schematic diagram showing the state of the flipped-out pages of the book as they pass through the thickness / pressure distribution sensing submodule of the present invention. Figure 7 This is a schematic diagram showing the state of the book pages when they are turned up and stabilized at an inclined angle.

[0021] In the diagram: 1. Cabinet; 2. Touchscreen transparent screen; 3. Book return slot; 4. Control unit; 5. Air blowing device; 51. Nozzle; 6. Conveyor roller assembly; 7. Flexible page turning mechanism; 71. Electric lifting twin-screw mechanism; 72. First motor; 73. Friction roller; 8. Infrared transmission scanning submodule; 81. Infrared linear light source; 82. High-resolution linear array camera; 9. Thickness / pressure distribution sensing submodule; 10. 3D topography scanning submodule; 101. Miniature structured light projector; 102. Camera; 11. Spectral reflectance analysis submodule; 12. Positioning and pushing device; 121. Second motor; 122. Lead screw; 123. Push plate. Detailed Implementation

[0022] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0023] In the description of this invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0024] See Figures 1-7 This invention provides a self-service book lending machine with a transparent touchscreen. The cabinet 1 is typically vertical, and a transparent touchscreen 2 is prominently displayed on the front of the cabinet 1, serving as the core interface for user interaction. The cabinet 1 houses a book storage and retrieval mechanism, including at least one borrowing slot and a return slot 3 with a return channel. The borrowing and return slots 3 can be combined or separate. The cabinet 1 contains hardware modules that implement various functions, all uniformly scheduled and controlled by a core control unit 4 (typically an industrial computer).

[0025] The core component of this invention is a detection module for multi-dimensional fusion intelligent detection of book integrity, which is integrated and installed in the book return channel. When the user puts the book to be returned into the book return port 3, the book is first received by a basic conveyor roller group 6 and guided into the detection station. After the detection is completed, the book is sent to the feeding station or returned to the book return port 3 according to the instructions (at this time, the sorting function is performed).

[0026] The positioning and pushing device 12 is driven by a linear mechanism to move the push plate 123. In this embodiment, the linear mechanism specifically uses a second motor 121 to drive the lead screw 122 to rotate, thereby causing the push plate 123 to move linearly. The push plate 123 covers the detection station and the feeding station. The feeding station is downstream of the detection station and leads to the book compartment inside the cabinet 1. At the detection station, the positioning and pushing device 12 triggers the positioning function, pushing the book towards the side of the return slot 3 (e.g., ...) using the push plate 123. Figure 4 The system locates the books; at the feeding station, if the books are found to be in good condition, the positioning and pushing device 12 triggers the sorting function and pushes the books to the book warehouse.

[0027] The flexible page-turning mechanism 7 can be a robotic arm with a flexible suction nozzle or friction head, or two sets of biomimetic rollers with slight clamping and frictional forces that can move alternately. Its core design principle is to move gently and adapt to books of varying thicknesses and paper materials. Its task is to unfold the book pages from a closed state, page by page or intermittently, at angles of approximately 10-30 degrees each time, so that subsequent sensors can scan them. The page-turning speed and angle are adaptively adjusted by the control unit 4 based on the book's thickness and number of pages.

[0028] In this embodiment, the flexible page-turning mechanism 7 specifically includes an electric lifting twin-screw mechanism 71 and a friction roller 73 driven by a first motor 72 and installed at the free end of the electric lifting twin-screw mechanism 71. When the friction roller 73 rotates in contact with the page, the free end of the page in direct contact with it will first form an arc shape (e.g., Figure 5 Then, the free end of the page will be flipped from below the friction roller 73 to above the friction roller 73. The electric lifting twin-screw mechanism 71 has a built-in displacement sensor. After the friction roller 73 completes a single page-turning action, the displacement sensor detects the change in the distance between the friction roller 73 and the next page, and then sends a trigger signal to the central processing module. The central processing module drives the twin screw to rotate, causing the friction roller 73 to move down precisely until the surface of the friction roller 73 is in contact with the next page. After the displacement sensor feeds back the contact signal, the twin screw stops running and waits for the next page-turning command. This cycle is used to achieve automatic continuous page turning.

[0029] The flexible page-turning mechanism 7 is paired with an air-blowing device 5 installed on one side of the book return channel. When the page is turned over above the friction roller 73, the air ejected from the inclined nozzle 51 on the air-blowing device 5 blows air onto the lower surface of the turned page, stabilizing the page at an inclined angle (e.g., ...). Figure 7 This process ensures that each subsequent page undergoes the same comprehensive scanning and inspection as the previous page. Once the inspection is complete and the blowing device 5 stops, the blown pages will return to their original positions under the influence of gravity.

[0030] Surrounding the flexible page-turning mechanism 7, a multi-sensor fusion detection array is precisely arranged. Each sub-module of the multi-sensor fusion detection array scans the unfolded pages synchronously or rapidly in sequence from different angles in a non-contact or micro-contact manner.

[0031] Infrared Transmission Scanning Submodule 8: This submodule is used to detect the macroscopic integrity of book pages, such as missing pages or large-area tears. It mainly consists of an infrared linear light source 81 and a high-resolution line scan camera 82. The infrared linear light source 81 is installed below (or to one side) the book page, emitting uniform infrared light upwards (or to the other side). The high-resolution line scan camera 82 is installed above the book page, directly facing the light source. When the book page is laid out between the two, infrared light penetrates the paper. Intact paper has a specific and uniform attenuation of infrared light. Once a page is missing (e.g., one or more pages are torn), the amount of light transmitted at that point increases sharply, forming an abnormally bright stripe or area in the image captured by the high-resolution line scan camera 82. The AI ​​model can easily identify this abrupt change by comparing the light transmission images of adjacent pages, thus determining that a page is missing. The principle is similar for large-area tears on the same page.

[0032] Thickness / Pressure Distribution Sensing Submodule 9: This submodule is used to detect microscopic physical defects in the paper, such as a torn corner, a cut hole, or foreign objects like sticky notes or leaves. It uses a flexible thin-film pressure distribution sensor, which is placed on a support bracket in the book return channel, shaped like a soft mat. When a page is turned and gently laid flat or brushed against the sensor surface (e.g., ...), the sensor detects these defects. Figure 6 The micro-pressure exerted by the paper on the sensor surface creates a detailed pressure distribution cloud map. On a healthy, uniformly thick sheet of paper, the pressure distribution is continuous and smooth; if a corner is torn, the pressure at that location drops sharply to zero (or near zero), creating a hollow distribution feature; if a foreign object is trapped, the pressure at that location will abnormally increase. By analyzing the real-time pressure distribution map obtained from each scan using an AI model and comparing it pixel-level with the standard pressure map of the page (from a scan of newly received books), millimeter-sized defects or foreign objects can be precisely located.

[0033] 3D Topography Scanning Submodule 10: This submodule is used to detect physical deformations on the paper surface, such as dents caused by forceful writing with a ballpoint pen or fountain pen, severe creases, and paper wrinkles or warping caused by water stains drying. It employs structured light projection technology, including a miniature structured light projector 101 and a matching camera 102. The miniature structured light projector 101 projects a series of coded grating stripes onto the open page surface. These stripes are regular on a flat paper surface, but will be distorted on dented or uneven paper surfaces. The camera 102 captures these distorted stripes from another angle. Through the principle of triangulation, the central processing module can reconstruct the 3D point cloud data of the page surface, thereby accurately quantifying the surface height changes. By analyzing the 3D topography data, the AI ​​model can determine whether there are abnormal dents or protrusions and classify them as "pen indentations," "creases," or "stain deformation."

[0034] Spectral Reflectance Analysis Submodule 11: This submodule analyzes changes in the chemical composition of the paper surface to identify handwriting (pencil, fountain pen, highlighter), oil stains, water stains, mold, etc. It mainly consists of a multispectral LED light source and a miniature spectrometer. The multispectral LED light source emits light of specific wavelengths, such as ultraviolet (UV), blue, green, and red light from the visible spectrum, and even near-infrared light. When these different wavelengths of light illuminate the page surface, different substances (such as wood pulp paper, carbon ink, dye ink, and grease) reflect unique spectral curves. The miniature spectrometer receives the reflected light and performs spectroscopic analysis to obtain the spectral "fingerprint" of the current irradiated point. By comparing it with the pre-stored spectral baseline of "clean paper," the AI ​​can determine with high confidence whether there are new notes in the area (and can initially distinguish the type of pen) or specific types of stains. This submodule typically works in conjunction with page-turning actions using point scanning or line scanning.

[0035] The raw data (images, pressure matrix, 3D point cloud, spectral curves) collected by all sensor submodules (8, 9, 10, 11) are uploaded to the central processing module in real time via a high-speed data bus. The central processing module is a dedicated computing unit built into the control unit 4 or working in conjunction with it. It is equipped with a pre-trained AI analysis model, which is a multi-task fusion neural network. Its training data includes a large number of calibrated multi-sensor data pairs of various types of book damage.

[0036] The specific inspection process when returning books is as follows: Step S301: The user initiates the book return operation through the touch-sensitive transparent screen 2 and places the book into the return slot 3 in the correct position according to the instructions.

[0037] Step S302: The basic RFID / barcode reader reads the book tag to complete identification. The control unit 4 retrieves the book's "standard digital file" stored in the cloud or locally. If it is the first return (i.e., no file), the system may prompt the user to enter "new book archiving mode" to perform a complete non-destructive scan to create a file.

[0038] Step S303: Control unit 4 starts the detection process, conveyor roller group 6 sends the book into the detection module, and flexible page turning mechanism 7 starts working.

[0039] Step S304: During the page turning process, the infrared transmission scanning submodule 8, the thickness / pressure distribution sensing submodule 9, the three-dimensional topography scanning submodule 10, and the spectral reflectance analysis submodule 11 are triggered synchronously according to a predetermined time sequence to collect data for each frame (or every few pages) of the book.

[0040] Step S305: The central processing module receives all real-time data and drives the AI ​​analysis model to work. The AI ​​analysis model does not process the data from each sensor independently, but performs feature-level fusion. For example, it performs correlation analysis and weighted decision-making on abnormal areas in infrared images, void features in pressure distribution maps, concave data of 3D morphology, and possible spectral changes, and finally determines whether damage exists, the type of damage (such as "missing top left corner of page 35, about 1 cm²", "continuous notes with water-based pen on pages 50-55", "crease on page 120"), and the level of damage.

[0041] Step S306: The central processing module generates the final diagnostic report.

[0042] Step S307: Human-machine interaction and sorting. The diagnostic report is sent to control unit 4, and control unit 4 performs the following operations: a) On the touch-sensitive transparent screen 2, an augmented reality (AR) technology is used to overlay a virtual book model with an animation of the detection process. For example, the virtual book will turn pages synchronously, and the area being scanned will be highlighted in real time. After the detection is completed, if the book is intact, a success message will be displayed; if the book is damaged, the problem will be clearly marked with icons (such as hole icons or pen icons) at the corresponding page number position in the virtual book in the form of animation, along with a text list explanation.

[0043] b) According to preset rules (such as "slight pencil marks do not affect reading and can be passed; missing pages need to be processed"), the control unit 4 instructs the positioning and pushing device 12 to operate and return the book to the storage, or instructs the transmission roller group 6 to operate and send the book out of the return slot 3.

[0044] c) Provide users with interactive buttons on the screen, such as "Confirm Result" or "Appeal Against Objection." The entire testing and feedback process should ideally be completed within 60 seconds to ensure a good user experience.

[0045] Through the above implementation methods, the present invention achieves automated, high-precision, and multi-dimensional detection of the integrity of returned book pages, significantly improving the intelligence level and efficiency of library asset management.

[0046] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A self-service book lending machine with a transparent touch screen, comprising a cabinet (1), a transparent touch screen (2) mounted on the cabinet (1), a control unit (4), and a book storage and retrieval mechanism disposed inside the cabinet (1), wherein the book storage and retrieval mechanism includes at least one borrowing slot and one returning slot (3), characterized in that: Inside the book return port (3), a multi-dimensional fusion intelligent detection module for detecting the integrity of books is integrated along the book delivery path; The detection module includes: A flexible page-turning mechanism (7) is used to automatically turn the pages during detection; A multi-sensor fusion detection array is arranged around the flexible page-turning mechanism (7) to collect multi-dimensional physical feature data of the turned pages; The central processing module is connected to the control unit (4) and has a pre-set AI analysis model. It is used to receive real-time data from the multi-sensor fusion detection array, compare and analyze the data with the pre-stored standard feature data, and generate a book page integrity diagnosis report. The touch-sensitive transparent screen (2) is configured to dynamically display the detection process and show the diagnostic report.

2. The self-service book lending machine with a transparent touchscreen as described in claim 1, characterized in that: The multi-sensor fusion detection array includes at least an infrared transmission scanning submodule (8), a thickness / pressure distribution sensing submodule (9), a three-dimensional topography scanning submodule (10), and a spectral reflectance analysis submodule (11).

3. A self-service book lending machine with a transparent touchscreen as described in claim 2, characterized in that: The infrared transmission scanning submodule (8) includes an infrared linear light source (81) and a high-resolution linear array camera (82), which are respectively set on both sides of the page transmission path and are used to detect missing pages and large-area tears by analyzing the difference in infrared light transmission.

4. A self-service book lending machine with a transparent touchscreen as described in claim 2, characterized in that: The thickness / pressure distribution sensing submodule (9) includes a flexible thin film pressure distribution sensor laid on the bearing surface of the book return channel, used to acquire the pressure distribution map of the book pages to detect local defects and foreign objects.

5. A self-service book lending machine with a transparent touchscreen as described in claim 2, characterized in that: The three-dimensional topography scanning submodule (10) includes a miniature structured light projector (101) and a camera (102) for reconstructing the three-dimensional topography of the page surface using structured light technology to detect dents and creases.

6. A self-service book lending machine with a transparent touchscreen as described in claim 2, characterized in that: The spectral reflectance analysis submodule (11) includes a multispectral LED light source and a miniature spectrometer, used to detect handwriting and stains by analyzing the reflectance spectral characteristics of the page surface.

7. A self-service book lending machine with a transparent touchscreen as described in claim 1, characterized in that: The AI ​​analysis model in the central processing module is a multi-task fusion neural network, which is used to perform feature-level fusion and correlation analysis on data from the multi-sensor fusion detection array.

8. A self-service book lending machine with a transparent touchscreen as described in claim 1, characterized in that: The touch-sensitive transparent screen (2) uses augmented reality technology to overlay the virtual book model with the real-time detection process animation, and makes a visual annotation at the corresponding position of the virtual model when damage is detected.

9. A self-service book lending machine with a transparent touchscreen according to any one of claims 1 to 8, characterized in that: The cabinet (1) is also equipped with a positioning and pushing device (12) controlled by the control unit (4). The positioning and pushing device (12) is driven by a linear mechanism to move the push plate (123). The positioning and pushing device (12) triggers the positioning function at the detection station according to the instruction, or triggers the sorting function together with the conveyor roller group (6) at the feeding station.

10. A self-service book lending machine with a touch-screen transparent screen according to any one of claims 1 to 8, characterized in that: It also includes an air blowing device (5) installed on one side of the book return channel for use with the flexible page turning mechanism (7). The air blowing device (5) stabilizes the turned pages at an inclined angle through the inclined nozzle (51).

Images