Contact type image acquisition device and acquisition processing method

Abstract

The invention discloses a contact type image acquisition device and an acquisition processing method, and belongs to the field of image acquisition devices and algorithms. The method comprises the steps that a displacement sensor detects the displacement in the x direction and the displacement in the y direction, and when the displacement reaches the scanning line width, a main control chip sends a starting signal to a time sequence controller; when an image time sequence is generated, data receiving and conversion are carried out, and an inertial measurement unit collects coordinate and attitude data; processing the image data, and displaying a processing result on a screen in real time; s2, continuously scanning, and repeatedly updating and processing the process in S1 in the scanning process; the image is displayed, subsequent processing is carried out, and the subsequent processing comprises printing, storing, editing and deleting; and image acquisition and processing are completed. According to the method and the device, the scanned image can be checked in real time, the scanning speed is high, the efficiency is remarkably improved, the scanning plane is not limited by the width of an image acquisition sensor, any plane image can be acquired in real time and displayed in real time, and the image scanning result can be observed conveniently.

Description

Technical Field

[0001] The present invention relates to the field of image acquisition devices and algorithms, and in particular to a contact image acquisition device and method. Background Art

[0002] Most existing scanning technologies are based on the uniform movement of stepper motors or the use of rollers to drive encoders to generate scanning signals. Scanning devices based on guide rails (scanners, fax machines) are large and inconvenient to move, and require computer host software to input images. For roller-based image scanning devices, conventional roller encoders have complex physical structures of gratings or magnetic gratings, and resolution cannot be taken into account, which makes them impossible to be lightweight and miniaturized. The movable parts of the rollers are easily damaged, resulting in a short service life of the device. For example, the patent CN202679477U device uses rollers and encoder triggers, which are large in size, the roller mechanism is easily damaged, and the scanning content cannot be viewed in real time, affecting efficiency.

[0003] Traditional track-type, roller-type and roller-type trigger scanning are difficult to make lightweight and miniaturized; in special scenarios, scanning cannot be completed due to the heavy device body; and the rollers and internal gears of traditional handheld scanning devices are subject to obvious external forces during handheld scanning contact, and their service life is uncontrollable. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies of the prior art and to provide a contact type image acquisition device and an acquisition processing method.

[0005] The objective of the present invention is achieved through the following technical solutions:

[0006] A contact image acquisition device, the device comprising: a displacement sensor, an inertial measurement unit, an image acquisition sensor, a data conversion chip, a timing controller, a main control chip, a display touch screen, a power supply, a memory and a wireless communication module;

[0007] Acquisition control and processing part: the main control chip is connected to the displacement sensor and inertial measurement unit, timing controller, data conversion chip, wireless communication module, display touch screen and memory; the image acquisition sensor is connected to the data conversion chip, and the image acquisition sensor and data conversion chip are connected to the timing controller;

[0008] The memory, display touch screen, wireless communication module and acquisition control and processing part are connected to a power source for power supply;

[0009] The displacement sensor includes: a photoelectric displacement sensor and a laser displacement sensor;

[0010] The inertial measurement unit includes: an acceleration sensor, a gyroscope, and a magnetometer.

[0011] In certain circumstances, the function of the timing controller can also be completed by the main control chip.

[0012] Under certain conditions (non-high-precision scanning line width), the displacement sensor can also be replaced by an inertial measurement unit, specifically acceleration measurement.

[0013] Furthermore, the displacement sensor may also be: a magnetoelectric displacement sensor, a resistive displacement sensor, a capacitive displacement sensor, and an ultrasonic displacement sensor.

[0014] A contact device image acquisition and processing method is provided, the method comprising the steps of:

[0015] S1. The displacement sensor detects the displacement in the x and y directions, and when the scan line width is reached, the main control chip sends a start command to the timing controller; while generating the image timing, data is received and converted, and the inertial measurement unit collects coordinate and attitude data; then the image data is processed and the processing results are displayed on the screen in real time;

[0016] S2. Continue scanning, and repeat the process of updating and processing S1 during the scanning process;

[0017] S3. Display the image and perform subsequent processing, including printing, saving, editing and deleting; complete the acquisition and processing of the image.

[0018] Furthermore, a panoramic scanning image is synthesized through the data collected by the displacement sensor and the inertial measurement unit;

[0019] The image data Image[] array of image P1 is obtained by scanning. The displacement sensor and the inertial measurement unit obtain the starting point coordinates P1 (x1, y1) and the relative rotation angle θ1 of the image when scanning. After the image is affected by the rotation angle, the normal image is rotated according to the corresponding angle;

[0020] Image rotation: Assume that any point (x, y) on the image is rotated counterclockwise by RotaryAngle angle around a coordinate point (rx0, ry0) to obtain a new coordinate (x′, y′), x is translated by rx0, and y is translated by ry0; the formula is:

[0021] x′=(x-rx0)*cos(RotaryAngle)+(y-ry0)*sin(RotaryAngle)+rx0;

[0022] y′=-(x-rx0)*sin(RotaryAngle)+(y-ry0)*cos(RotaryAngle)+ry0;

[0023] The formula for finding the source coordinate point based on the new coordinate point is:

[0024] x=(x′-rx0)*cos(RotaryAngle)-(y′-ry0)*sin(RotaryAngle)+rx0;

[0025] y=(x′-rx0)*sin(RotaryAngle)+(y′-ry0)*cos(RotaryAngle)+ry0;

[0026] The image array is rotated by the corresponding RotaryAngle angle to form Image1[], and the corresponding coordinates of the center point of the rotated image are calculated, P1′(x1, y1);

[0027] The images P1, P2, P3, ..., Pn are processed through the above steps, and finally all the images are combined into a panoramic image according to the coordinate values ​​of the images.

[0028] Furthermore, the acceleration detected by the inertial measurement unit is , then the velocity signal obtained after one time domain integration is:

[0029] ;

[0030] The displacement signal obtained after two time domain integrations is:

[0031] ;

[0032] in, and are the initial velocity and initial displacement respectively, and are the theoretical velocity and displacement to be determined, and is the original function of velocity and displacement.

[0033] Furthermore, the synthesized panoramic image is subjected to noise reduction:

[0034] Denoising is based on wavelet transform, which combines images into different frequency bands. The wavelet representation is:

[0035] ;

[0036] in is the wavelet coefficient, is the wavelet basis function.

[0037] Wavelet transform is a signal processing technique used to represent an image as a combination of different frequency bands to extract and remove noise. By controlling the wavelet coefficients, noise can be reduced and a clearer image can be reconstructed.

[0038] Furthermore, after the panoramic image is denoised, it is repaired by a restoration model, and the restoration model is:

[0039] ;

[0040] in, is the frequency response of the image, is the noise power, is the power spectral density of the original image. Image restoration aims to reconstruct the original image from a blurred or damaged image. The Wiener filter is a commonly used method that achieves this by minimizing the mean square error in the frequency domain.

[0041] The beneficial effects of the present invention are:

[0042] (1) Using displacement sensors or inertial measurement units as scan triggers and scan control signals effectively reduces the size of the device, reduces the complexity of the hardware structure, and reduces the maintenance cost of the device;

[0043] (2) Displacement sensors, especially photoelectric displacement sensors, have high acquisition accuracy and are easy to control in software. Different acquisition intervals can be configured according to different scanning objects.

[0044] (3) Use displacement sensors to trigger scanning signals, replace traditional tracks or rollers, adopt new image acquisition modules, and efficient image generation algorithms. Use displacement sensors and inertial measurement units to synthesize any collected data into a panoramic scanning image, and can collect images of surfaces of any size. Real-time viewing of scanned images, fast scanning speed, and significantly improved efficiency. The scanning plane is not limited by the width of the image acquisition sensor. Any plane image can be collected and displayed in real time to facilitate observation of the scanning quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a main structural diagram of a contact image acquisition device according to this embodiment.

[0047] Figure 2 It is a flow chart of the main steps of the present invention;

[0048] Figure 3 This is a schematic diagram of picture rotation in this embodiment;

[0049] Figure 4 This is the process diagram of P1 angle conversion;

[0050] Figure 5 It is the merged image of images P1~P5; DETAILED DESCRIPTION

[0052] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0053] See also Figure 1 , provides a contact image acquisition device, characterized in that the device includes: a displacement sensor, an inertial measurement unit, an image acquisition sensor, a data conversion chip, a timing controller, a main control chip, a display touch screen, a power supply, a memory and a wireless communication module;

[0054] Acquisition control and processing part: the main control chip is connected to the displacement sensor, inertial measurement unit, timing controller, data conversion chip, wireless communication module, display touch screen and memory; the image acquisition sensor is connected to the data conversion chip, and the image acquisition sensor and data conversion chip are connected to the timing controller;

[0055] The memory, display touch screen, wireless communication module and acquisition control and processing part are respectively connected to power supplies for power supply:

[0056] The displacement sensor includes: a photoelectric displacement sensor, a laser displacement sensor, a magnetoelectric displacement sensor, a resistive displacement sensor, a capacitive displacement sensor, and an ultrasonic displacement sensor.

[0057] The inertial measurement unit includes: an acceleration sensor, a gyroscope, and a magnetometer.

[0058] In certain circumstances, the function of the timing controller can also be completed by the main control chip.

[0059] Under certain conditions (non-high-precision scanning line width), the displacement sensor can also be replaced by an inertial measurement unit, specifically acceleration measurement.

[0060] The main control chip generates complete image information of the scanned surface based on the recorded scan coordinates, displacement sensor, and image data obtained by the image sensor, and displays it in real time on the screen. A high-speed data bus is used to transmit data between the ADC chip and the main control chip. The main control chip receives data in DMA mode, and saves the scan coordinates (x, y) and the attitude information of the inertial measurement unit while receiving the image data. When the device scans the data to the minimum data frame, it performs image correction calculation. Different from the traditional data acquisition method that needs to be transmitted through USB or FGPA, this method has low cost, high speed and adapts to the data acquisition rate.

[0061] The image acquisition sensor is responsible for converting the spectrum signal of the acquisition surface into a voltage signal. The data conversion chip converts the voltage signal of the sensor into a digital signal. The main control chip directly reads the digital signal. The timing controller is used to provide high-frequency control timing signals to the image acquisition sensor and the data conversion chip. At the same time, it provides a synchronous clock signal to the main control chip and the data conversion chip to achieve high-speed data reading.

[0062] The displacement sensor and the inertial measurement unit are directly connected to the main control chip. The displacement sensor is responsible for scanning and moving displacement detection of the device, and the inertial measurement unit records the relative position and posture of the current device.

[0063] Each time the device moves a scanning line, the main control chip triggers a scanning task to the timing controller. After completing a scanning action, the image sensor and data conversion chip enter the waiting state. When the next scanning task arrives, the timing controller starts the image sensor and data conversion module to perform the next scanning task.

[0064] The main control chip performs color balance calculation in real time every time it receives a line of image data, and records the scanned x, y displacement data of the image. The relative position of the current scanned line is determined based on the scan origin. At the same time, the three-dimensional spatial relationship between the current scan coordinates and the scan origin can be determined based on the inertial measurement unit. Image reconstruction of the three-dimensional space plane can be achieved through calculation processing.

[0065] The color balance is calculated as:

[0066] 1) The image acquisition sensor converts the color component into an electrical signal. Since there will be a slight deviation in the data during the photoelectric conversion of each sensor, the deviation of each pixel is obtained by averaging multiple times;

[0067] 2) When the device leaves the factory, it scans multiple lines (s) through the white balance and black balance planes, each line has (l) pixels. This continuous data is stored in a two-dimensional digital buffer[] (size lxs).

[0068] 3) After scanning, calculate the average value of each pixel.

[0069] Pavg0=Averagea(buffer[0][0], buffer[1][0]..., buffer[s-1][0]);

[0070] Pavg1=Averagea(buffer[0][1], buffer[1][1]..., buffer[s-1][1]);

[0071] Pavgl-1=Averagea(buffer[0][l-1],buffer[1][l-1]...,buffer[s-1][l-1]);

[0072] At the same time, Pavg0 Pavg1 ...Pavgl-1 are saved in the system; each time a frame of image is completed, a color calibration is completed for each pixel using the above average value. The blackboard balance adjustment algorithm is:

[0073] Vy=Pixsize*(Vx-PVd) / (PVp-PVd)+K;

[0074] Vy is the pixel value after compensation, Vx is the pixel value to be compensated, PVp is the average pixel value of the white card, PVd is the average pixel value of the black card, Pixsize is the maximum value of the single channel color, 8bit is 255, 10bit is 1024, and K is the compensation value.

[0075] As shown in the architecture diagram: Scanning starts, the system starts the main control chip and configures the scanning accuracy (according to the characteristics of the image acquisition sensor, an interrupt is generated every 0.8mm movement, which is the most suitable application scenario). The handheld device moves on the scanning plane, and the main control chip generates an interrupt every time the moving distance reaches (0.8mm). The main control chip receives the interrupt and starts to read the displacement x and y values ​​to record the current displacement. If x is a negative number n (n is a negative number), the net movement value d=n is recorded. d<0 means that the image acquisition sensor moves in the reverse direction and the system does not trigger scanning. The next interrupt x is a positive m (m is an integer), indicating that the image acquisition sensor moves in the forward direction. The net positive movement value d=d+m. If m+n>0, it means that the net positive value is greater than 0. d>0, and a row scan signal can be generated at this time (at the same time d=0). According to the parameter configuration of the main control chip, ideally m and n can only be 1, that is, scanning will only be triggered when the net increase value of the image acquisition sensor moving in the forward direction is greater than 0.

[0076] The main control chip triggers a high level to the timing controller. The timing controller receives the image clock signal generated by the main control chip, drives the data conversion chip and the image acquisition sensor to work, and the main control chip reads the ADC data through DMA according to the data clock generated by the timing chip.

[0077] After the main control chip reads a complete line of data, it records the coordinates (x′, y′) of this scan and the scanning angle θ′ of the scanning device, and finally forms a complete image frame by scanning multiple lines of data.

[0078] Using photoelectric displacement sensors and inertial measurement units as scanning triggers and scanning control signals effectively reduces the size of the device. It reduces the complexity of the hardware structure and reduces the maintenance cost of the device. Semiconductor displacement sensors represented by photoelectric displacement sensors

[0079] The triggering accuracy is high, the software control is simple, and different acquisition intervals can be configured according to different scanning objects. Real-time viewing of scanned images, fast scanning speed, and significantly improved efficiency. The scanning plane is not limited by the width of the image sensor, and any plane image can be collected in real time and displayed in real time for easy observation of the scanning quality.

[0080] Example 2

[0081] See also Figure 2 , provides a contact device image acquisition and processing method, the method steps comprising:

[0082] S1. The photoelectric displacement sensor detects the displacement in the x and y directions, and when the scan line width is reached, the main control chip sends a start command to the timing controller; while generating the image timing, data is received and converted, and the inertial measurement unit collects coordinate and attitude data; then the image data is processed and the processing results are displayed on the screen in real time;

[0083] S2. Continue scanning, and repeat the process of updating and processing S1 during the scanning process;

[0084] S3. Display the image and perform subsequent processing, including printing, saving, editing and deleting; complete the acquisition and processing of the image.

[0085] See also Figure 2-Figure 3 , a panoramic scanning image is synthesized through the data collected by the displacement sensor and the inertial measurement unit;

[0086] The image data Image[] array of image P1 is obtained by scanning. The displacement sensor and the inertial measurement unit obtain the starting point coordinates P1 (x1, y1) and the relative rotation angle θ1 of the image when scanning. After the image is affected by the rotation angle, the normal image is rotated according to the corresponding angle;

[0087] See also Figure 3 , Image rotation: Assume that any point (x, y) on the image is rotated counterclockwise by the RotaryAngle angle around a coordinate point (rx0, ry0) to obtain a new coordinate (x′, y′), x is translated by rx0, y is translated by ry0, and the RotaryAngle angle corresponding to angle a is:

[0088] x′=(x-rx0)*cos(RotaryAngle)+(y-ry0)*sin(RotaryAngle)+rx0;

[0089] y′=-(x-rx0)*sin(RotaryAngle)+(y-ry0)*cos(RotaryAngle)+ry0;

[0090] The formula for finding the source coordinate point based on the new coordinate point is:

[0091] x=(x′-rx0)*cos(RotaryAngle)-(y′-ry0)*sin(RotaryAngle)+rx0;

[0092] y=(x′-rx0)*sin(RotaryAngle)+(y′-ry0)*cos(RotaryAngle)+ry0;

[0093] See also Figure 4 The image array is rotated by the corresponding RotaryAngle angle to form Image1[], and the corresponding coordinates of the center point of the rotated image are calculated, P1′(x1, y1);

[0094] See also Figure 5 , Figure 5 The combined image of P1 to P5. Through the above steps, images P1, P2, P3...Pn are processed, and finally all images are combined into a panoramic image according to the coordinate values ​​of the images.

[0095] The acceleration detected by the inertial measurement unit is , then the velocity signal obtained after one time domain integration is:

[0096] ;

[0097] The displacement signal obtained after two time domain integrations is:

[0098] ;

[0099] in, and are the initial velocity and initial displacement respectively, and are the theoretical velocity and displacement to be determined, and is the original function of velocity and displacement.

[0100] Wavelet transform is a signal processing technique used to represent an image as a combination of different frequency bands to extract and remove noise. By controlling the wavelet coefficients, noise can be reduced and a clearer image can be reconstructed.

[0101] Denoising is based on wavelet transform, which combines images into different frequency bands. The wavelet representation is:

[0102] ;

[0103] in, is the panoramic image to be processed, is the wavelet coefficient, is the wavelet basis function. Use wavelet transform to process panoramic images , get the wavelet coefficients, assuming that after wavelet transform:

[0104] ;

[0105] Setting the Threshold :

[0106] After denoising:

[0107] ;

[0108] in, < ,so , through inverse transformation, a clearer image is restored.

[0109] Image restoration aims to reconstruct the original image from a blurred or damaged image. The Wiener filter is a commonly used method that is implemented by minimizing the mean square error in the frequency domain. After the panoramic image is denoised, it is repaired using a restoration model, which is:

[0110] ;

[0111] in, is the frequency response of the image, is the noise power, is the power spectral density of the original image. Assume that the blurring process of an image is performed by a known PSF The noise power is measured as , assuming :Then we have:

[0112] ;

[0113] Then the Wiener filter is:

[0114] ;

[0115] The image is then restored after filtering, and finally a panoramic scanning image obtained during the contact process is obtained.

[0116] This embodiment effectively reduces the size of the device by using a displacement sensor and an inertial measurement unit as a scan trigger and a scan control signal. The hardware structure complexity is reduced, and the device maintenance cost is low. Semiconductor displacement sensors represented by photoelectric displacement sensors have high triggering accuracy and simple software control. Different acquisition intervals can be configured according to different scanning plane parameters. The scanned image can be viewed in real time, the scanning speed is fast, and the efficiency is significantly improved. The scanning plane is not limited by the width of the image acquisition sensor, and any plane image can be acquired in real time and displayed in real time to facilitate observation of the scanning quality.

[0117] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art shall not deviate from the spirit and scope of the present invention, and shall be within the scope of protection of the claims attached to the present invention.

Claims

1. A contact image acquisition device, characterized in that: The device includes: a displacement sensor, an inertial measurement unit, an image acquisition sensor, a data conversion chip, a timing controller, a main control chip, a display touch screen, a power supply, a memory and a wireless communication module; Acquisition control and processing part: the main control chip is connected to the displacement sensor and inertial measurement unit, timing controller, data conversion chip, wireless communication module, display touch screen and memory; the image acquisition sensor is connected to the data conversion chip, and the image acquisition sensor and data conversion chip are connected to the timing controller; The memory, display touch screen, wireless communication module and acquisition control and processing part are connected to a power source for power supply; The displacement sensor includes: a photoelectric displacement sensor and a laser displacement sensor; The inertial measurement unit includes: an acceleration sensor, a gyroscope, and a magnetometer.

2. A contact image acquisition device according to claim 1, characterized in that: The displacement sensor may also be a magnetoelectric displacement sensor, a resistive displacement sensor, a capacitive displacement sensor, or an ultrasonic displacement sensor.

3. A contact device image acquisition and processing method, characterized in that: The method steps include: S1. The displacement sensor detects the displacement in the x and y directions, and when the scan line width is reached, the main control chip sends a start command to the timing controller; while generating the image timing, data is received and converted, and the inertial measurement unit collects coordinate and attitude data; then the image data is processed and the processing results are displayed on the screen in real time; S2. Continue scanning, and repeat the process of updating and processing S1 during the scanning process; S3. Display the image and perform subsequent processing, including printing, saving, editing and deleting; complete the acquisition and processing of the image.

4. The method for image acquisition and processing of a contact device according to claim 3, characterized in that: A panoramic scanning image is synthesized through the data collected by the displacement sensor and the inertial measurement unit; The image data Image[] array of image P1 is obtained by scanning. The displacement sensor and the inertial measurement unit obtain the starting point coordinates P1 (x1, y1) and the relative rotation angle θ1 during the image scanning. Image rotation: Assume that any point (x, y) on the image is rotated counterclockwise by RotaryAngle angle around a coordinate point (rx0, ry0) to obtain a new coordinate (x′, y′), x is translated by rx0, and y is translated by ry0; the formula is: x′=(x-rx0)*cos(RotaryAngle)+(y-ry0)*sin(RotaryAngle)+rx0; y′=-(x-rx0)*sin(RotaryAngle)+(y-ry0)*cos(RotaryAngle)+ry0; The formula for finding the source coordinate point based on the new coordinate point is: x=(x′-rx0)*cos(RotaryAngle)-(y′-ry0)*sin(RotaryAngle)+rx0; y=(x′-rx0)*sin(RotaryAngle)+(y′-ry0)*cos(RotaryAngle)+ry0; The image array is rotated by the corresponding RotaryAngle angle to form Image1[], and the corresponding coordinates of the center point of the rotated image are calculated, P1′(x1, y1); The images P1, P2, P3, ..., Pn are processed through the above steps, and finally all the images are combined into a panoramic image according to the coordinate values ​​of the images.

5. The method for image acquisition and processing of a contact device according to claim 3, characterized in that: The acceleration detected by the inertial measurement unit is , then the velocity signal obtained after one time domain integration is: ； The displacement signal obtained after two time domain integrations is: ； in, and are the initial velocity and initial displacement respectively, and are the theoretical velocity and displacement to be determined, and is the original function of velocity and displacement.

6. The method for collecting and processing images of a contact device according to claim 3, characterized in that: It also includes denoising the synthesized panoramic image: Denoising is based on wavelet transform, which combines images into different frequency bands. The wavelet representation is: ； in is the wavelet coefficient, is the wavelet basis function.

7. The method for collecting and processing images of a contact device according to claim 5, characterized in that: After the panoramic image is denoised, it is repaired using a restoration model, and the restoration model is: ； in, is the frequency response of the image, is the noise power, is the power spectral density of the original image.

Images