Calligraphy and painting cultural relic technology photographing system

Abstract

The invention provides a technical photographing system for calligraphy and painting cultural relics, and relates to the technical field of photographing and repairing of the calligraphy and painting cultural relics, and the system comprises an automatic light source adjustment unit which comprises a light source recognition module and a light source optimization module, and the light source recognition module is used for recognizing physical feature parameters of the calligraphy and painting cultural relics and optimizing the light source parameters of the calligraphy and painting cultural relics; the physical characteristic parameters comprise materials, sizes and surface characteristics of the calligraphy and painting cultural relics. According to the photographing system for the calligraphy, painting and cultural relic technology, a multi-spectral image sequence is cooperatively collected by combining multiple light sources, detailed data including an ultraviolet fluorescence image, an infrared reflection image, a light-transmitting image and a sidelight texture image can be obtained, and the light sources in the lamp set are automatically selected and combined; therefore, precise illumination adjustment for different types of calligraphy and painting cultural relics is realized, the acquisition quality of the multispectral image sequence is ensured, and the accuracy and definition of the image are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of shooting and repairing of calligraphy and cultural relics, in particular to a technical photography system for calligraphy and cultural relics. BACKGROUND

[0002] At present, the photography technology of calligraphy and cultural relics has been widely used in the field of cultural relic protection and repair. The traditional shooting method mainly relies on single light source and fixed shooting system, using conventional cameras combined with halogen lamps or white light lamps and other basic lighting equipment. Although this kind of system can complete basic image acquisition, it is difficult to adapt to the diversity of different calligraphy and cultural relics in material quality, texture and preservation state. The existing technology generally lacks the ability of multi-spectral collaborative shooting, and cannot synchronously obtain key characteristic data such as ultraviolet fluorescence and infrared reflection, resulting in incomplete record of cultural relic information. For special materials such as silk and rice paper, the traditional lighting method is easy to produce mirror reflection, which seriously affects the detail restoration accuracy.

[0003] The core problem of the existing system is the lack of intelligence, which cannot automatically optimize the shooting parameters according to the characteristics of cultural relics. Moreover, the existing equipment is mostly fixed structure, which is difficult to adapt to the shooting needs of cultural relics of different sizes or types, and lacks systematic historical data management function, resulting in that the repair process cannot refer to the optimization parameters of similar cultural relics. Especially in the aspect of light source control, there is a lack of adaptive adjustment mechanism based on material identification, which is difficult to balance the lighting needs of different wave bands. The image processing link relies on manual repair, which is insufficient in restoring the accuracy of fine cracks and faded areas, and the repair results lack historical data support. This mechanical working mode leads to unstable imaging quality, and it is difficult to meet the high-precision digitization needs of precious cultural relics. SUMMARY

[0004] In view of the deficiencies of the prior art, the present application provides a technical photography system for calligraphy and cultural relics, which solves the problems of poor image quality, improper light source selection and insufficient repair accuracy in traditional calligraphy and cultural relic shooting through multi-spectral image acquisition, automatic light source adjustment and high-precision image processing repair process.

[0005] To achieve the above purpose, the technical photography system for calligraphy and cultural relics is realized by the following technical scheme: a technical photography system for calligraphy and cultural relics, comprising: A fixed frame, the upper surface of the fixed frame is fixedly connected with a shooting table, the upper surface of the shooting table is fixedly connected with a strip-shaped LED lamp, the upper surface of the strip-shaped LED lamp is fixedly connected with an ultraviolet light source halogen lamp, the strip-shaped LED lamps are distributed on the left and right sides, the upper side of the shooting table and between the strip-shaped LED lamps is fixedly connected with a gantry, and the lower side of the gantry is fixedly connected with a shooting assembly; A flash support platform is fixedly connected to the side surface of the mounting bracket, and a flash is fixedly connected to the upper side of the flash support platform. A diffuser layer is fixedly connected to the side surface of the mounting bracket and the upper side of the flash support platform. There is a gap between the diffuser layer and the shooting table surface. The shooting components, the strip LED light, and the flash are all connected to the outside. The light assembly includes the ultraviolet light source halogen lamp, the flash lamp, and the strip industrial LED lamp. The light assembly and the imaging component work together to acquire a multispectral image sequence, which includes ultraviolet fluorescence images, infrared reflection images, light transmission images, and side-lit texture images. An automatic light source adjustment unit is provided, comprising a light source recognition module and a light source optimization module. The light source recognition module is used to identify the physical characteristic parameters of the calligraphy and painting artifacts, including the material, size, and surface characteristics of the calligraphy and painting artifacts.

[0006] Preferably, the gantry frame is used to support and fix the shooting assembly, and the shooting platform is used to place the calligraphy and painting artifacts to be photographed.

[0007] Preferably, the automatic light source adjustment unit is used to automatically select and combine the light groups according to the physical characteristic parameters of the calligraphy and painting artifacts.

[0008] Preferably, the ultraviolet halogen lamp is used to provide ultraviolet and infrared illumination, the flash lamp is used to generate transmitted light and infrared transmitted light, and the strip LED lamp is used to generate lateral textured illumination.

[0009] Preferably, the image processing and virtual restoration module uses wavelet transform thresholding and least squares method to perform denoising and color correction on the multispectral image sequence. The specific operation of wavelet transform thresholding is as follows: .

[0010] in, The denoised multispectral image sequence is... The wavelet transform matrix is... These are the wavelet transform coefficients at various scales. The first of the multispectral image sequence Data points, The total number of image data points, with the noise standard deviation controlled below 0.02; The specific operation of the least squares method is as follows: .

[0011] in, For color conversion matrix, To correct the pixel values ​​of the image before, is a pixel value in the standard color reference image, is the total number of pixels, the objective of the least square method is to minimize the difference between the images before and after correction, ensuring that the color difference does not exceed 2.5.

[0012] Preferably, the image processing and virtual repair module stores the image of the calligraphy and painting cultural relics in TIFF format.

[0003] Preferably, the image processing and virtual repair module generates a three-dimensional model according to the stored image, the three-dimensional model can present the surface texture, color and micro-defects of the calligraphy and painting cultural relics, the three-dimensional model is repaired by a virtual repair algorithm, and the specific formula of the virtual algorithm is: .

[0014] wherein, is the pixel value after repair, is the pixel value of the neighborhood to be repaired, is the repair weight function, is the radius of the repair window, and is the coordinate of the current pixel, the formula restores the missing or damaged part of the image by adding the neighborhood pixel value, and the repair accuracy is within 0.1mm.

[0015] Preferably, the soft light layer comprises light diffusion material and replaceable filter, which is used to homogenize the light distribution and suppress the specular reflection.

[0016] The present application provides a kind of calligraphy and painting cultural relics technical photography system. With the following beneficial effects: The calligraphy and painting cultural relics technical photography system, by combining a variety of light sources cooperates and acquires multispectral image sequence, can obtain detailed data including ultraviolet fluorescence image, infrared reflection image, light transmission image and side light texture image, and automatically selects and combines the light source in lamp group, to realize accurate light adjustment for different types of calligraphy and painting cultural relics, ensure the acquisition quality of multispectral image sequence, improve the accuracy and clarity of image.

[0017] Adopt multispectral image processing and virtual repair module, combine wavelet transform threshold and least square method for denoising and color correction, so that the calligraphy and painting cultural relics image obtained is clearer and has high color restoration degree. By generating a three-dimensional model and performing virtual repair, the micro-defects in the image are further repaired, the accuracy and repair quality of the image are improved, and the repair accuracy is ensured to be within 0.1mm. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 is a schematic diagram of the internal structure of the implementation of the application; Figure 2 This is a schematic diagram of a system structure for implementing an invention; Figure 3 This is a schematic diagram showing the configuration and functional breakdown of a lamp assembly that realizes the invention; Figure 4 This is a flowchart illustrating the image processing and virtual repair process used in this invention. Figure 5 This is a flowchart of an automatic light source adjustment unit that implements the invention; Figure 6 This is a schematic diagram of the overall structure for realizing an invention; Figure 7 It is a way to realize an invention. Figure 6 Enlarged schematic diagram of the structure at point A in the middle; Figure 8 This is another overall structural diagram for realizing the invention.

[0019] The components include: 1. Mounting frame; 2. Flash support platform; 3. Flash unit; 4. Strip LED light; 5. Softening layer; 6. Shooting table; 7. Ultraviolet halogen lamp; 8. Gantry; and 9. Shooting components. Detailed Implementation

[0020] 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. Example 1

[0021] like Figures 1-8 As shown, this embodiment of the invention provides a technical photography system for calligraphy and painting cultural relics, including: A fixed frame 1 is provided, with a shooting platform 6 fixedly connected to its upper surface. The shooting platform 6 measures 120cm x 150cm and is used to place the calligraphy and painting artifacts to be photographed. A strip LED light 4, 120cm in length, is fixedly connected to the upper surface of the shooting platform 6. An ultraviolet halogen lamp 7, 15cm in diameter, is fixedly connected to the upper surface of the strip LED light 4. The strip LED lights 4 are distributed on the left and right sides. A gantry frame 8 is fixedly connected to the upper side of the shooting platform 6 and between the strip LED lights 4. A shooting component 9 is fixedly connected to the lower side of the gantry frame 8. The gantry frame 8 is used to support and fix the shooting component 9. The shooting platform 6 is used to place the calligraphy and painting artifacts to be photographed.

[0022] A flash support platform 2 is fixedly connected to the side surface of the mounting bracket 1. A flash unit 3 is fixedly connected to the upper side of the flash support platform 2. A diffuser layer 5 is fixedly connected to the side surface of the mounting bracket 1, located above the flash support platform 2. There is a gap between the diffuser layer 5 and the shooting table surface 6. The shooting assembly 9, the strip LED light 4, and the flash unit 3 are all connected to the outside. The diffuser layer 5 contains light-diffusing material and replaceable filters, used to homogenize light distribution and suppress specular reflections.

[0023] The lighting assembly includes a UV halogen lamp 7, a flash 3, and a strip industrial LED light 4. The lighting assembly and the imaging component 9 work together to acquire a multispectral image sequence, which includes UV fluorescence images, infrared reflectance images, transmitted light images, and side-lit texture images. The UV halogen lamp 7 provides UV and infrared illumination, the flash 3 generates transmitted light and infrared transmitted light, and the strip LED light 4 produces side-lit texture illumination.

[0024] The image processing and virtual restoration module uses wavelet transform thresholding and least squares method to perform denoising and color correction on multispectral image sequences. The specific operation of wavelet transform thresholding is as follows: .

[0025] in, This is the denoised multispectral image sequence. The wavelet transform matrix is... These are the wavelet transform coefficients at various scales. The first multispectral image sequence Data points, The total number of image data points, with the noise standard deviation controlled below 0.02; The specific operation of the least squares method is as follows: .

[0026] in, For color conversion matrix, To correct the pixel values ​​of the image before, These are the pixel values ​​in the standard color reference image. Given the total number of pixels, the goal of the least squares method is to minimize the difference between the images before and after correction, ensuring that the color difference does not exceed 2.5.

[0027] The image processing and virtual restoration module stores images of calligraphy and painting artifacts in TIFF format.

[0028] The image processing and virtual restoration module generates a 3D model based on the stored image. This 3D model can represent the surface texture, color, and minor defects of the calligraphy and painting artifacts. The 3D model is then used to restore the image through a virtual restoration algorithm. The specific formula for this virtual algorithm is as follows: .

[0029] wherein, is the repaired pixel value, is the neighborhood pixel value of the image to be repaired, is the repair weight function, is the radius of the repair window, and is the coordinate of the current pixel, the formula recovers the missing or damaged part of the image by adding the neighborhood pixel value, and the repair accuracy is within 0.1 mm.

[0030] The automatic light source adjusting unit comprises a light source identification module and a light source optimization module. The light source identification module is used for identifying physical characteristic parameters of the calligraphy and painting cultural relics. The physical characteristic parameters include the material, size and surface characteristics of the calligraphy and painting cultural relics. The automatic light source adjusting unit is used for automatically selecting and combining lamp groups according to the physical characteristic parameters of the calligraphy and painting cultural relics.

[0031] The fixed frame and the shooting table form a rigid "optical darkroom" skeleton, which suppresses environmental vibration and stray light, and ensures pixel-level alignment when long exposure or high magnification shooting is performed. The left and right symmetrical arrangement of the strip-shaped LED lamps provides visible light, eliminates the mirror reflection on the surface of the calligraphy and painting, and provides uniform basic lighting for subsequent three-dimensional modeling. The ultraviolet light source halogen lamp is stacked on the LED to make the optical axes of the ultraviolet light and the visible light same, thereby reducing the parallax caused by multiple clamping. The continuous spectrum of the halogen lamp can excite various fluorescent substances in the 200-400nm segment, thereby improving the information abundance. The gantry forms a "door" shaped unobstructed light path in the center, which is convenient for mechanical arm or manual turning, and also serves as the installation reference of the multi-axis camera sliding rail.

[0032] The external flash lamp bearing platform physically isolates the transient high-power power supply, capacitor and precise shooting components, thereby reducing electromagnetic interference and thermal drift. The flash lamp and the soft light layer, the flash lamp provides a millisecond high-energy pulse, which solves the problem of low signal-to-noise ratio caused by insufficient ultraviolet light flux. The soft light layer converts the point light source into a >120° diffuse area light source, which suppresses the overflow of high light and retains the texture of paper and silk. The soft light layer and the table leave a gap to form an "air vibration isolation" and heat dissipation channel, which prevents the temperature rise caused by continuous flashing from being ≥2℃, thereby avoiding the virtual focus caused by thermal expansion and cold movement.

[0033] The ultraviolet fluorescence image detects the organic fluorophores of the repair glue, mold spots and seal of the seal, and locates the pen repair area. The infrared reflection image in the 900-1700nm wave band penetrates the pigment layer, and reveals the covered draft and inscription. The transmission image reads the paper fiber direction, watermark and crack distribution in the backlight mode, and is used for paper aging evaluation. The side light texture image: 30° low angle incidence, strengthens the concave-convex of the pen mark, generates a 3D micro-topography with a precision of 1μm, and assists in judging the pen force and the difference of the repair touch feeling.

[0034] The light source identification module uses a multi-spectral reflectance curve to distinguish silk, rice paper, and leather paper, and automatically matches the optimal ultraviolet dose to avoid light aging. Laser or structured light ranging is used to adjust the LED current and flash energy in real time, so that A4-A0 formats can all achieve 18% gray card standard exposure. Polarized imaging is used to detect the surface gloss, and if the gloss is greater than 40, a cross-polarizer is automatically enabled to reduce mirror reflection.

[0035] The light source optimization module calculates in real time according to the SNR, adjusts the flash pulse width and LED duty cycle, and ensures that the gray scale utilization rate is greater than 90% for 12-bit color depth. When the LED substrate temperature is greater than 45°C, the power is automatically reduced and the fan is triggered to maintain a wavelength drift of less than 2nm, ensuring color consistency.

[0036] Example Two Unlike Example One, the application scenario of this embodiment is the non-destructive, high-precision digital protection and copying of calligraphy and cultural relics. The specific implementation is as follows: The fixed frame, shooting table, and gantry are integrally formed from high-rigidity alloy, with a whole-body resonance frequency of more than 50Hz, ensuring that the mechanical drift is less than 0.01mm during long-term exposure. The shooting table surface is covered with replaceable acid-free linen, which prevents the cultural relics from sliding and avoids chemical pollution. The longitudinal and transverse slides of the gantry are both equipped with magnetic grid displacement feedback, with a repeat positioning accuracy of ±0.02mm, ensuring that the shooting assembly maintains the same optical axis during multiple scans. The shooting assembly is designed to be adjustable in height, with a lifting stroke of 200mm to accommodate the focal plane adjustment of different thickness spools or vertical shafts. The superimposed height of the strip-shaped LED lamp and the ultraviolet halogen lamp is simulated optically to maintain a 30° angle between the centers of the two lamps and the normal line of the cultural relic surface, providing uniform illumination and reducing direct glare.

[0037] A honeycomb aluminum damping pad is provided between the flash lamp bearing platform and the fixed frame, which can absorb 90% of the flash lamp reaction force within 1ms, avoiding image ghosting caused by micro-vibration. The soft light layer is composed of a double-layer micro-structured diffuser plate and a magnetic filter frame, with a haze value of the diffuser plate of ≥85%, and a pluggable UV / IR cutoff filter, which can achieve an energy uniformity of 0.1% within the 200nm-1100nm wavelength band. The 30mm gap between the soft light layer and the shooting table forms a controllable airflow channel, which maintains a laminar flow state through a silent fan to remove the heat generated by the flash lamp within 3s, preventing the surface temperature of the cultural relic from rising by more than 1°C.

[0038] The UV light source halogen lamp adopts a double-filament design, which can quickly switch between 254 nm-400 nm and 700 nm-1000 nm. The inner wall of the lampshade is coated with deep ultraviolet high-reflection film, making the irradiance at the center wavelength of 365 nm reach 3 mW / cm², meeting the signal-to-noise ratio requirements of fluorescence excitation. The flash lamp is configured with a double-pulse mode, with the first pulse used for transmission imaging and the second pulse delayed by 5 ms to trigger infrared transmission, which enhances the contrast of fiber structures inside the film or silk by 40%. The strip-shaped industrial LED lamp realizes 0-100% continuous adjustment through pulse width modulation technology, and the side light incidence angle can be steplessly changed between 15°-60° to highlight the height difference of strokes. The depth resolution of the side light texture image reaches 5µm.

[0039] The wavelet basis function is selected as a function with 5 layers of decomposition. The soft threshold strategy is combined with adaptive estimation to process high-frequency textures and low-frequency color blocks separately, which eliminates random noise and preserves stroke details. The color correction process introduces a 3x3 linear matrix based on IT8.7 / 2 color card and a 1D lookup table cascade structure, which is iteratively optimized in the ∆E color space. The final color difference is better than the threshold value that can be recognized by the human eye. The built-in GPU acceleration pipeline in the module processes a single 16-bit TIFF format 100 million pixel image in less than 3 seconds, meeting the batch scanning requirements.

[0040] The uncompressed TIFF6.0 specification is used, with 16 bits per channel, supporting XMP metadata embedding, recording light source parameters, exposure time, lens model, and temperature and humidity sensor data, ensuring scientific-level repeatability. The file naming follows the ISO19115 standard, automatically includes UUID to avoid duplication or coverage. The storage path is backed up by RAID-6 array and LTO-9 tape to ensure 30 years of data integrity.

[0041] The three-dimensional reconstruction adopts a structure light and photometric stereo fusion scheme, the point cloud density is ≥1000 pts / mm2, and the surface normal error is <0.5°. During the structure light three-dimensional reconstruction process, the system supports a hybrid encoding technology combining multi-frequency phase shift method and Gray code, which can achieve sub-pixel level accuracy in stripe decoding on complex surface textures, effectively avoiding point cloud defects caused by high reflection areas or uneven ink color. The Poisson surface reconstruction algorithm is introduced in the point cloud post-processing stage to ensure the continuity and smoothness of the model, while preserving the true geometric features at the cracks and worm-eaten places. The photometric stereo module supports multi-directional light source array switching, combined with polarization imaging technology, effectively improving the detail capture capability of low reflectivity areas. The light intensity control of the structure light projection unit is below 0.5 mW / cm2, which is much lower than the safe threshold of cultural relics, ensuring that there is no light damage during long-time scanning. The weight function in the virtual restoration algorithm combines the geodesic distance and color similarity, and the adaptive window radius r varies between 3-15 pixels, synthesizing the texture of worm-eaten, cracks, and mold spots, and the gradient continuity of the edge of the missing area is maintained above 95%. The restoration results are output as high-fidelity 8K texture maps and PBR materials, which can be directly used for digital exhibitions or 3D printing replication.

[0042] The light source recognition module integrates a hyperspectral camera and a laser confocal sensor to complete non-contact material identification within 3s, with a classification accuracy of >98% for paper, silk, silk, ink, and mineral pigments. The size measurement adopts binocular grating projection, with a thickness error of ±0.05 mm and a surface roughness Ra measurement range of 0.1-100 µm. The light source optimization module searches for the optimal illumination distribution in 2¹² lamp combinations in real time through a genetic algorithm, with a target function considering signal-to-noise ratio, heat, and exposure time. The typical optimization time is <0.5s, ensuring that each shot obtains the highest image quality within the safe threshold of cultural relics.

[0043] Through the above steps, the book and painting cultural relic technical photography system meets or exceeds the scientific research level standard in key indicators such as resolution, color accuracy, temperature rise control, vibration suppression, and data security, achieving high-fidelity, non-destructive, and repeatable digital acquisition and three-dimensional reconstruction of book and painting cultural relics.

[0044] Example Three This example is based on a book and painting cultural relic technical photography system, which verifies whether the automatic light source adjustment unit can complete light source combination decision-making and output multi-spectral images with color difference meeting the system requirement of not more than 2.5 according to the identified material, size, and surface reflectivity in real book and painting cultural relic shooting scenes. The specific implementation is as follows: 1. Pretreatment Lay the Qing Dynasty silk scroll Set Color Vertical Shaft “Autumn Mountain Traveling Picture” to be shot in the center of the shooting table 6, and wipe the surface dust with a dust-free cloth.

[0045] 2. Light source recognition The system light source recognition module reads the physical characteristics of the cultural relics through the built-in sensor: Material determination: silk scroll.

[0046] Actual size: 132.4 cm long, 65.7 cm wide.

[0047] Surface reflectivity: average value 37%.

[0048] 3. Light source combination decision.

[0049] The light source optimization module automatically schedules the light group according to the above characteristics: The bar-shaped LED light 4 is adjusted to 55% of the nominal value, providing lateral texture illumination.

[0050] The ultraviolet light source halogen lamp 7 emits at a wavelength of 360 nm, with a current setting of 2.6 A.

[0051] The flash light 3 is set to a single output energy of 35 J, with an angle self-adjustment of 30° from the vertical axis.

[0052] 4. Multi-spectral sequence acquisition The shooting assembly 9 is driven by the gantry 8 to continuously shoot in the order of "visible light-ultraviolet fluorescence-infrared reflection-transmitted light", a total of 4 images are obtained.

[0053] 5. Data verification An independent color card is used to sample synchronously at the same location, and the color difference is measured , which meets the system internal color difference ≤2.5 determination threshold.

[0054] This example completes the multi-spectral image acquisition of "Autumn Mountain Travel Map" through a complete automatic light source adjustment process without human intervention, with the measured color consistency, verifying that the automatic light source adjustment unit has good self-adaptive ability for different materials, sizes and surface reflection characteristics of calligraphy and painting cultural relics.

[0055] Example Four This example is based on the denoising function of the image processing and virtual restoration module of a calligraphy and painting cultural relics technical photography system, and performs wavelet transform threshold denoising on the multi-spectral image sequence of calligraphy and painting cultural relics. The specific implementation is as follows: 1. Data preparation Input data: collected ultraviolet fluorescence image sequence of silk scroll landscape painting in TIFF format, size 512x512 pixels, total data points .

[0056] Noise evaluation: the standard deviation of the noise of the original image is 0.035 by calculating the gray value of the uniform area of the image.

[0057] 2. Wavelet transform denoising operation The image is decomposed into three levels using a classical wavelet basis for multi-scale image decomposition to obtain the coefficients at each scale. .

[0058] The specific operation of wavelet transform thresholding is as follows: .

[0059] in, This is the denoised multispectral image sequence. The wavelet transform matrix is... These are the wavelet transform coefficients at various scales. The first multispectral image sequence Data points, The total number of image data points, with the noise standard deviation controlled below 0.02.

[0060] For the first data point The pixel value at coordinate (1,1) is 128. High frequency coefficient , .

[0061] Mid-frequency coefficient , .

[0062] Low frequency coefficient , .

[0063] Weighted sum .

[0064] Inverse Transformation Reconstruct pixel values .

[0065] The calculation is completed by traversing all data points.

[0066] 3. Noise control The standard deviation of noise in the uniform region of the output image is reduced to 0.018, which meets the requirement of <0.02.

[0067] Key results: Table 1: Noise Control Results.

[0068] Parameter Original image De-noised image Noise standard deviation 0.035 0.018 Edge sharpness Blur Significant boost 4. Data storage The denoised ultraviolet fluorescence image is stored in TIFF format.

[0069] This example demonstrates the effectiveness of the system in denoising UV fluorescence images of calligraphy and cultural relics by processing 262,144 image data points using wavelet transform thresholding, reducing the noise standard deviation from 0.035 to 0.018. The denoised images significantly improve the recognition accuracy of surface cracks and mineral pigments, providing a high-quality data foundation for subsequent virtual restoration.

[0070] Example Four This example is based on a technical photography system for calligraphy and cultural relics, demonstrating how to use the multispectral imaging technology in the system to obtain UV fluorescence images of calligraphy and cultural relics, and repair the images through image processing technology. The specific implementation is as follows: 1. Equipment configuration Shooting table 6: size 100cm x 100cm, used to place the calligraphy and cultural relics to be photographed.

[0071] UV light source halogen lamp 7: power 50W, wavelength range 300nm-400nm, used to provide UV illumination.

[0072] Image acquisition component 9: equipped with a high-resolution camera with a resolution of 4000x3000 pixels, capable of capturing high-detail UV fluorescence images.

[0073] Soft light layer 5: thickness 2mm, using light diffusion material to ensure uniform light.

[0074] 2. Light source configuration Start the UV light source halogen lamp 7, which outputs UV light with a wavelength range of 300nm-400nm, to excite the fluorescence reaction on the surface of calligraphy and cultural relics.

[0075] After light irradiation, the surface of the cultural relics produces different intensity of UV fluorescence.

[0076] The distance between the light source and the surface of the cultural relics is kept at 50cm to ensure sufficient light intensity.

[0077] 3. Image acquisition Use the image acquisition component 9 to take pictures, set the camera resolution to 4000x3000 pixels, and the total number of image pixels is 12 million.

[0078] Set the camera exposure time to 1 / 250 seconds, aperture to F8, and ISO sensitivity to 400 to ensure that the captured image avoids overexposure and loss of details.

[0079] 4. Image denoising processing Assume that the image noise standard deviation is 0.02. Use wavelet transform for denoising processing, and its mathematical model can be expressed as: .

[0080] wherein, is the denoised image, is the wavelet transform matrix, is the inverse wavelet transform matrix. For simplicity, assume the matrix size is 10x10 and the noise standard deviation is 0.02.

[0081] Specific processing example: Assume that the original value of a pixel in the image is about 80, and the actual observation value fluctuates between 78 and 82 under the interference of noise with a standard deviation of 0.02. After wavelet transform and denoising processing, the pixel value can be restored to 79.5, which is closer to the true value.

[0082] 5. Image restoration: The image processing and virtual restoration module generates a three-dimensional model based on the stored image. The model can present the surface texture, color, and minor defects of the painting and cultural relics. The image is restored by a virtual restoration algorithm, and the algorithm formula is defined as: .

[0083] wherein, is the restored pixel value, is the neighborhood pixel value of the image to be restored, is the restoration weight function, is the radius of the restoration window, and is the coordinate of the current pixel. The formula restores the missing or damaged part of the image by adding the neighborhood pixel values, and the restoration accuracy is within 0.1 mm.

[0084] Restoration process example: Take the restoration of pixel (100, 150) as an example. The neighborhood pixel coordinates are (99, 149), (100, 149), (101, 150), (99, 150), and (100, 151), and the corresponding pixel values are 70, 75, 80, 90, and 85, respectively.

[0085] Set the weight function Based on the similarity calculation of the neighborhood pixel and the target pixel value, for example, it can be defined as: .

[0086] The greater the weight, the greater the contribution of the neighborhood pixel to the restoration value. The restoration calculation process for pixel (100, 150) is as follows: Calculate the weight of each neighborhood pixel: .

[0087] .

[0088] .

[0089] .

[0090] .

[0091] The repaired pixel value is: .

[0092] .

[0093] The repaired pixel value is 121.28 by weighted average.

[0094] 6. Storage and analysis: The processed ultraviolet fluorescence image is stored in TIFF format, and the size of a single file is about 40 MB.

[0095] The processed image data can generate a three-dimensional model with a precision of 0.1 mm, which can accurately reproduce the surface texture and micro-defects of the painting and cultural relics.

[0096] In summary, through the operation process, the ultraviolet fluorescence image of the painting and cultural relics can be accurately obtained, and the wavelet transform is used for noise suppression and the virtual repair technology is used for image restoration. The repaired image can restore the details of the cultural relics with high fidelity, especially in the reconstruction of micro-defects and surface texture, and the repair precision reaches 0.1 mm, which provides high-quality image data basis for the repair and analysis of cultural relics.

[0097] Although embodiments of the present application have been shown and described, it is to be understood that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the present application, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A calligraphy and cultural relic technical photography system, characterized in that, The utility model relates to a kind of book and painting cultural relics photographing device, including: Fixed frame (1), the upper surface of the fixed frame (1) is fixedly connected with shooting mesa (6), the upper surface of the shooting mesa (6) is fixedly connected with strip LED lamp (4), the upper surface of the strip LED lamp (4) is fixedly connected with ultraviolet light source halogen lamp (7), the strip LED lamp (4) is distributed left and right, the upper side of the shooting mesa (6) and between strip LED lamp (4) is fixedly connected with gantry (8), the lower side of the gantry (8) is fixedly connected with shooting assembly (9); The side surface of the fixed frame (1) is fixedly connected with flash lamp bearing platform (2), the upper side of the flash lamp bearing platform (2) is fixedly connected with flash lamp (3), the side surface of the fixed frame (1) and on the upper side of the flash lamp bearing platform (2) is fixedly connected with soft light layer (5), there is gap between the soft light layer (5) and shooting mesa (6), the shooting assembly (9), strip LED lamp (4), flash lamp (3) are connected with outside; Lamp group, the lamp group includes the ultraviolet light source halogen lamp (7), the flash lamp (3) and the strip industrial LED lamp (4), the lamp group cooperates with the shooting assembly (9) and collects multispectral image sequence, the multispectral image sequence includes ultraviolet fluorescence image, infrared reflection image, light transmission image and side light texture image; Automatic light source adjusting unit, the automatic light source adjusting unit includes light source identification module and light source optimization module, the light source identification module is used to identify the physical characteristic parameter of the book and painting cultural relics, the physical characteristic parameter includes the material, size and surface characteristic of the book and painting cultural relics; Image processing and virtual repair module is used to process the multispectral image sequence.

2. The calligraphy and cultural relic technical photography system according to claim 1, characterized in that: The gantry (8) is used to support and fix the shooting assembly (9), and the shooting mesa (6) is used to place the book and painting cultural relics to be photographed.

3. The calligraphy and cultural relic technical photography system according to claim 1, characterized in that: The automatic light source adjusting unit is used to automatically select and combine the lamp group according to the physical characteristic parameter of the book and painting cultural relics.

4. The calligraphy and cultural relic technical photography system according to claim 1, characterized in that: The ultraviolet light source halogen lamp (7) is used to provide ultraviolet and infrared waveband illumination, the flash lamp (3) is used to generate transmission light and infrared transmission light, and the strip LED lamp (4) is used to generate lateral texture illumination.

5. The calligraphy and cultural relic technical photography system according to claim 1, characterized in that: The image processing and virtual repair module adopts wavelet transform threshold and least square method to carry out denoising processing and color correction processing to the multispectral image sequence, and the specific operation of the wavelet transform threshold is: ， in, The denoised multispectral image sequence is... The wavelet transform matrix is... These are the wavelet transform coefficients at various scales. The first of the multispectral image sequence Data points, This represents the total number of image data points. The specific operation of the least square method is: ， wherein, is a color conversion matrix, is a pixel value of the pre-correction image, is a pixel value in the standard color reference image, is the total number of pixels.

6. The calligraphy and cultural relic technical photography system according to claim 1, characterized in that: The image processing and virtual repair module stores the image of the book and painting cultural relics in TIFF format.

7. The calligraphy and cultural relic technical photography system according to claim 6, characterized in that: The image processing and virtual repair module generates a three-dimensional model according to the stored image, and the three-dimensional model repairs the image through a virtual repair algorithm, and the specific formula of the virtual algorithm is: ， wherein, is the repaired pixel value, is the neighborhood pixel value of the image to be repaired, is the repair weight function, is the radius of the repair window, and is the coordinate of the current pixel.

8. The calligraphy and cultural relic technical photography system according to claim 1, characterized in that: The soft light layer (5) includes light diffusion material and replaceable filter.

Images