A polarization imaging system and method based on OCR technology

Abstract

The application discloses a kind of polarization imaging system and method based on OCR technology, the system includes incident light transmission imaging module, polarization mark module, mark character recognition module, polarization angle calculation module, polarization mark module is placed on the object plane of measured object, make the obtained image have the information of measured object and polarization mark simultaneously by incident light transmission imaging module, using mark character recognition module to identify and locate the polarization mark on the obtained image, the polarization angle of the obtained image is calculated according to the result of mark character recognition module using polarization angle calculation module, so as to realize complete polarization imaging process.The system designed by the application realizes the measurement of unknown polarization angle (but within the coverage) incident light by intelligently identifying the polarization mark on the object plane of measured object, the system structure is simple and real-time is higher, mobile terminal, lightweight device can be transplanted.

Description

Technical Field

[0001] This invention belongs to the field of optical imaging technology, specifically a polarization imaging system and method based on OCR technology. Background Technology

[0002] In polarization imaging technology, the polarization modulation system is an essential component. In active imaging systems, based on different modulation methods, it can be mainly divided into two types: mechanical rotation modulation and phase delay modulation. Mechanical rotation modulation mainly uses mechanical structures, such as motors, to control the rotation of polarizers or waveplates at a constant angular velocity to achieve polarization modulation of incident light. This method is simple to manufacture, low in cost, and has high accuracy, making it used in many commercial rotating analyzers. However, the measurement cycle is long, and mechanical vibrations can introduce errors, reducing accuracy. Phase delay modulation mainly uses electro-optic modulation. Liquid crystal adjustable phase delay devices (LCDs) have lower driving voltages, thus meeting the higher control voltage requirements of electro-optic modulators. This method uses the phase delay function of the LCD to polarize the incident light. It has fast modulation speed, strong real-time performance, and no moving parts, but it is more expensive, susceptible to temperature effects, and requires calibration and compensation. To reduce system costs and meet the market demand for portability and miniaturization, the system should avoid using additional hardware such as LCD adjustable phase delay devices, which would increase the system's hardware structure. At the same time, it should also avoid the problems of poor real-time performance and high error of traditional mechanical rotation modulation. Summary of the Invention

[0003] The purpose of this invention is to propose a polarization imaging system based on OCR technology.

[0004] The technical solution to achieve the purpose of this invention is as follows: a polarization imaging system based on OCR technology, comprising: an incident light transmission imaging module, a polarization marking module, a marking character recognition module, and a polarization angle calculation module. The polarization marking module is set on the surface of the object being measured so that the marking module can characterize the polarization angle of the current incident light. The incident light transmission imaging module includes a light source, a quarter-wave plate, a rotatable polarizer, and an imaging sensor area array camera. The light source is used to emit linearly polarized light onto the surface of the object being measured. After the linearly polarized light is reflected from the surface of the object being measured, it is polarized and modulated by the quarter-wave plate and the rotatable polarizer. A polarization image is obtained by the imaging sensor area array camera. The marking character recognition module is used to identify and locate the polarization markings on the obtained image. An automatic discrimination algorithm is used to determine whether the marking recognition is accurate. Based on the fixed positional relationship of the polarization marking modules, the positions of all polarization markings are calculated from a correctly identified marking position. The polarization angle calculation module calculates the polarization angle of the obtained image by curve fitting using the least squares method.

[0005] Preferably, the polarization marking module is a polarization mark fixed on the surface of the object being measured. The polarization mark is made of a material that is sensitive to light intensity information and will change in brightness according to the change of the polarization state of the incident light. Each mark corresponds to a polarization angle.

[0006] Preferably, the light source is an LED light source, arranged in a ring lighting pattern, and the lighting scheme is direct bright field front lighting.

[0007] Preferably, the fast axis of the quarter-wave plate is in the XY plane (with the incident light transmission direction as the X-axis and the vertical upward direction as the Y-axis) and forms a 45-degree angle with the X-axis, while the fast axis of the linear polarizer is along the X-axis.

[0008] Preferably, the light intensity of the acquired polarized light intensity image is specifically:

[0009]

[0010] In the formula, I is the average light intensity of the image, θ is the polarization angle of the image, I0 is the initial light intensity, φ is the linear delay, and δ is the azimuth angle.

[0011] Preferably, the identification character recognition module identifies and locates polarization markers on the acquired image, determines the accuracy of marker recognition through an automatic discrimination algorithm, and calculates the positions of all polarization markers based on the fixed positional relationship of the polarization marker module using a correctly identified marker position. The specific process is as follows:

[0012] Single character recognition: The method used is template matching, which involves scanning and matching a pre-made identifier template on the image and selecting the region with the best matching result as the identifier region.

[0013] Successful recognition judgment: The matching result is judged by calculating the Hamming distance between the matching region and the template image using the mean hash algorithm. If the Hamming distance is less than a set threshold, the recognition is successful.

[0014] Deducing all characters from a single character: By recognizing a single identifier, the positions of all identifiers in a region are calculated based on fixed spatial relationships. That is, a character in the polarization identifier on the input image is recognized, and it is determined whether the character recognition is correct. The position of the region identifier is calculated based on the correctly recognized character.

[0015] Preferably, the polarization angle calculation module calculates the polarization angle of the current image by performing curve fitting based on the light intensity information on the polarization marker in the obtained image and the polarization angle corresponding to the marker itself. The curve fitting equation is as follows:

[0016] f(x)=a+b cos(2x+c),x∈[0,π]

[0017] Where a, b, and c are coefficients, x is the polarization angle θ, and f(x) is the light intensity I of the image.

[0018] This invention also proposes a polarization imaging method based on OCR technology, the specific steps of which are as follows:

[0019] Step 1: In the incident light transmission imaging module, the polarization modulation of the incident light is achieved by rotating and controlling the rotatable polarizer, thereby obtaining a polarization image that simultaneously contains the object under test and polarization identification information.

[0020] Step 2: The character recognition module receives the initially acquired polarization image, uses OCR technology to identify and locate individual polarization markers, judges the accuracy of the recognition based on the automatic discrimination algorithm, and calculates the position of all polarization markers in the image based on the fixed settings of the polarization marker module.

[0021] Step 3: The polarization angle calculation module processes the average light intensity at the location of the polarization marker in the image and calculates the polarization angle of the polarization pattern by curve fitting.

[0022] Compared with existing technologies, the significant advantages of this invention are as follows: This invention does not require specially designed liquid crystal adjustable phase delay devices, temperature compensation modules, or complex conditions such as structured light illumination. It only uses traditional mechanical rotation modulation combined with OCR software technology to quickly modulate the incident light, realizing the measurement of incident light with unknown polarization angles (but within the coverage area); This invention can be used to manufacture lightweight, miniaturized, and portable polarization imaging devices; The software and hardware structure of this invention is simple to implement, and it can be mass-produced when combined with a mobile platform.

[0023] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0024] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.

[0025] Figure 1 This is a system structure diagram of the present invention.

[0026] Figure 2 This is the result image after the identifier is recognized on the PC.

[0027] Figure 3It is the result of the fitting curve of polarization angle and light intensity calculated from multiple input images. The curve conforms to the relationship between image light intensity and polarization angle of the incident light transmission imaging module. Detailed Implementation

[0028] It is readily understood that, based on the technical solution of this invention, various embodiments of the invention can be conceived by those skilled in the art without altering the essential spirit of the invention. Therefore, the following detailed embodiments and accompanying drawings are merely illustrative examples of the technical solution of this invention and should not be considered as the entirety of the invention or as limitations or restrictions on the technical solution of this invention. Rather, these embodiments are provided to enable those skilled in the art to gain a more thorough understanding of the invention. Preferred embodiments of the invention are described below in conjunction with the accompanying drawings, which form part of this application and, together with the embodiments of the invention, serve to illustrate the innovative concept of the invention.

[0029] The present invention is conceived as follows: Figure 1 As shown, a polarization imaging system based on OCR technology mainly achieves the measurement of incident light with unknown polarization angles (but within the coverage area) by intelligently recognizing polarization marks on the surface of the object under test. The system has a simple structure and high real-time performance, and the error caused by mechanical vibration is reduced through software algorithms. Specifically, it includes an incident light transmission imaging module, a polarization mark module, a mark character recognition module, and a polarization angle calculation module. The polarization mark module is placed on the surface of the object under test to represent the polarization angle of the current incident light. The incident light transmission imaging module includes a light source, a quarter-wave plate, a rotatable polarizer, and an imaging sensor area array camera. The light source emits linearly polarized light, which is reflected from the surface of the object under test. Polarization modulation is achieved through the quarter-wave plate and the rotatable polarizer, and the final polarization image is obtained through the camera. The mark character recognition module identifies and locates the polarization marks on the obtained image. An automatic discrimination algorithm is used to determine whether the mark recognition is accurate. Based on the fixed positional relationship of the polarization mark module, the positions of all polarization marks can be calculated from a correctly recognized mark position. The polarization angle calculation module calculates the polarization angle of the obtained image by curve fitting using the least squares method.

[0030] In a further embodiment, the polarization marking module is a polarization mark fixed to the surface of the object being measured. The polarization mark is made of a material sensitive to light intensity information, such as white paper, and will change in brightness according to the change of the polarization state of the incident light. Each mark corresponds to a polarization angle.

[0031] In a further embodiment, the incident light transmission imaging module includes a ring light source, a quarter-wave plate, a linear polarizer, and an imaging sensor array camera. The light signal reflected from the sample reaches the imaging sensor array camera via the quarter-wave plate and the linear polarizer, and the imaging sensor array camera obtains a polarization parameter image.

[0032] In a further embodiment, the light source is an LED light source, arranged in a ring lighting pattern, and the lighting scheme is direct bright field front lighting.

[0033] In a further embodiment, the fast axis of the quarter-wave plate is in the XY plane and forms a 45-degree angle with the X-axis, while the fast axis of the linear polarizer is along the X-axis.

[0034] In a further embodiment, the light intensity of the acquired polarized light intensity image is specifically:

[0035]

[0036] I represents the average light intensity of the image, and θ represents the polarization angle of the image.

[0037] In a further embodiment, the character recognition module includes a single character recognition process, a recognition success judgment process, and a single character derivation of all characters process. It recognizes a character in the polarization mark on the input image, judges whether the character recognition is correct, and calculates the position of the region mark based on the correctly recognized character.

[0038] In a further embodiment, the polarization angle calculation module can calculate the polarization angle of the current image by performing curve fitting based on the light intensity information on the polarization marker in the obtained image and the polarization angle corresponding to the marker itself. The curve fitting equation is as follows:

[0039] f(x(=a+b cos(2x+c),x∈[0,π]

[0040] Where a, b, and c are coefficients, x is the polarization angle θ, and f(x) is the light intensity I of the image.

[0041] The working process of this invention is as follows: A polarization marker is placed on the surface of the object being tested. A ring light source is incident on the object through a ring-shaped linear polarizer. A polarization marker sensitive to optical information is fixed on the surface of the object. After reflection, the incident light carries both the information of the object being tested and the optical marker. Then, the polarization state is modulated by passing through a quarter-wave plate and a polarizer. The polarization state of the incident light can be changed by rotating the polarizer. The modulated incident light is finally imaged on the target surface of the CCD through a lens. The CCD then sends the imaging information to the marker recognition module. The marker character recognition module identifies and locates the polarization markers on the acquired image. An automatic discrimination algorithm determines whether the marker recognition is accurate. Based on the fixed positional relationship of the polarization marker module, the positions of all polarization markers can be calculated from a correctly identified marker position. The polarization angle calculation module calculates the polarization angle of the acquired image by curve fitting using the least squares method.

[0042] A polarization imaging method based on OCR technology, the specific steps of which are as follows:

[0043] Step 1. In the incident light transmission imaging module, the polarization modulation of the incident light is achieved by rotating and controlling the rotatable polarizer, thereby obtaining a polarization image that simultaneously contains the object under test and polarization identification information.

[0044] Step 2. The character recognition module receives the initially acquired polarization image, uses OCR technology to identify and locate individual polarization markers, judges the accuracy of the recognition based on the automatic discrimination algorithm, and calculates the position of all polarization markers in the image based on the fixed settings of the polarization marker module.

[0045] In a further example, OCR technology is used to identify and locate polarization markers. One template matching method is introduced here: first, a template image is prepared, then the template image is placed on the image to be matched and slid across it. The similarity between the template image and the matching region is calculated using the squared difference matching algorithm. The formula for the squared difference matching method is as follows:

[0046]

[0047] In a further example, the Hamming distance between the template image and the matching region is calculated using the mean hash algorithm to determine whether the template matching is successful. If the Hamming distance is less than 10, the matching is considered successful. Based on the fixed relationship of the polarization identification module, the positions of all identifications are calculated from a successfully identified identification.

[0048] Step 3. The polarization angle calculation module processes the average light intensity at the location of the polarization marker in the image and calculates the polarization angle of the polarization pattern by curve fitting.

[0049] Specifically, the least squares method can be used to calculate the fitting curve between the marked angle and the average light intensity value in the marked area of ​​an image, determine the relationship between light intensity and angle, and find the angle corresponding to the position with the maximum light intensity value. This angle is the polarization angle of the input image. The calculated polarization angle conforms to the relationship between light intensity and polarization angle of the incident light transmission imaging module.

[0050] This invention uses traditional mechanical rotation modulation combined with OCR software technology to rapidly modulate incident light, enabling the measurement of incident light with unknown polarization angles (but within the coverage area). This invention can be used to manufacture lightweight, miniaturized, and portable polarization imaging devices. The software and hardware structure of this invention is simple to implement, and it can be mass-produced when combined with a mobile platform.

[0051] 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 changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

[0052] It should be understood that, in order to simplify the present invention and help those skilled in the art understand its various aspects, in the above description of exemplary embodiments of the present invention, various features of the present invention are sometimes described in a single embodiment or with reference to a single figure. However, the present invention should not be construed as implying that all features included in the exemplary embodiments are essential technical features of the claims of this patent.

[0053] It should be understood that the modules, units, components, etc., included in the device of one embodiment of the present invention can be adaptively changed to be placed in a device different from that embodiment. Different modules, units, or components included in the device of the embodiment can be combined into a single module, unit, or component, or they can be divided into multiple sub-modules, sub-units, or sub-components.

Claims

1. A polarization imaging system based on OCR technology, characterized in that, include: The system comprises an incident light transmission imaging module, a polarization marking module, a marking character recognition module, and a polarization angle calculation module. The polarization marking module is positioned on the surface of the object under test, allowing it to characterize the polarization angle of the current incident light. The incident light transmission imaging module includes a light source, a quarter-wave plate, a rotatable polarizer, and an imaging sensor area array camera. The light source emits linearly polarized light onto the surface of the object under test. After reflection, the linearly polarized light undergoes polarization modulation through the quarter-wave plate and the rotatable polarizer, and a polarization image is obtained through the imaging sensor area array camera. The marking character recognition module identifies and locates the polarization markings on the acquired image. An automatic discrimination algorithm determines the accuracy of the marking recognition. Based on the fixed positional relationship of the polarization marking modules, the positions of all polarization markings are calculated from a correctly identified marking position. The specific process is as follows: Single character recognition: The method used is template matching, which involves scanning and matching a pre-made identifier template on the image and selecting the region with the best matching result as the identifier region. Successful recognition judgment: The matching result is judged by calculating the Hamming distance between the matching region and the template image using the mean hash algorithm. If the Hamming distance is less than a set threshold, the recognition is successful. Deducing all characters from a single character: By recognizing a single identifier, the positions of all identifiers in a region are calculated based on fixed spatial relationships. That is, a character in the polarization identifier on the input image is recognized, and it is determined whether the character is correctly recognized. The position of the region identifier is calculated based on the correctly recognized character. The polarization angle calculation module calculates the polarization angle of the obtained image by curve fitting using the least squares method. The curve fitting equation is: in , , For coefficients, Polarization angle , The light intensity of the image .

2. The polarization imaging system based on OCR technology according to claim 1, characterized in that, The polarization marking module is a polarization mark fixed to the surface of the object being measured. The polarization mark is made of a material that is sensitive to light intensity information and will change in brightness according to the change of the polarization state of the incident light. Each mark corresponds to a polarization angle.

3. The polarization imaging system based on OCR technology according to claim 1, characterized in that, The light source is an LED light source, arranged in a ring lighting pattern, and the lighting scheme is direct bright field front lighting.

4. The polarization imaging system based on OCR technology according to claim 1, characterized in that, The fast axis of a quarter-wave plate lies in the XY plane and forms a 45-degree angle with the X-axis. The fast axis of a linear polarizer is along the X-axis. The XY plane has the direction of incident light propagation as the X-axis and the vertical upward direction as the Y-axis.

5. The polarization imaging system based on OCR technology according to claim 1, characterized in that, The specific light intensity of the acquired polarized light intensity image is as follows: In the formula, I is the average light intensity value of the image, and the polarization angle of the image is... I0 is the initial light intensity. For linear delay, This is the azimuth angle.

6. A polarization imaging method based on the system described in any one of claims 1 to 5, characterized in that, The specific steps are as follows: Step 1: In the incident light transmission imaging module, the polarization modulation of the incident light is achieved by rotating and controlling the rotatable polarizer, thereby obtaining a polarization image that simultaneously contains the object under test and polarization identification information. Step 2: The character recognition module receives the initially acquired polarization image, uses OCR technology to identify and locate individual polarization markers, judges the accuracy of the recognition based on the automatic discrimination algorithm, and calculates the position of all polarization markers in the image based on the fixed settings of the polarization marker module. Step 3: The polarization angle calculation module processes the average light intensity at the location of the polarization marker in the image and calculates the polarization angle of the polarization pattern by curve fitting.

Images