Non-uniformity correction method based on two-dimensional material linear array detection imaging system

Abstract

The invention discloses a heterogeneity correction method based on a two-dimensional material linear array detection imaging system. The heterogeneity correction method comprises the steps that S1, the voltage magnification times of all channels are adjusted according to light currents; s2, eliminating the amplified dark current by adopting a differential mode according to the amplification factor; s3, through linear array scanning, two-point calibration parameters are obtained; and S4, correcting the output of the detector according to the calibration parameters calibrated at the two points. The original output voltage of each detector is corrected using the two-point corrected gain coefficient and bias coefficient. According to the method, the gain and bias of each detector are adjusted on a hardware level, so that the consistency preliminary correction of photoelectric response is realized; and then a calibration algorithm is introduced in a software level, and the residual non-uniformity error is subjected to fine compensation, so that the imaging consistency and the image quality of the system are effectively improved, and the correction effect of a traditional method in practical application is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of detector imaging, and particularly relates to a non-uniformity correction method applied to a two-dimensional material linear array detection imaging system. Background Art

[0002] A two-dimensional material linear array detector is a photoelectric array detector constructed based on two-dimensional materials such as graphene and molybdenum disulfide. Its device structure is linearly arranged and can efficiently convert optical signals into electrical signals. This type of device has characteristics such as wide-spectrum detection, high sensitivity, fast response, and flexibility and bendability, and has advantages such as low power consumption, small size, and high performance. It is widely used in fields such as detection imaging, medical detection, environmental monitoring, and flexible electronics, and is an important development direction of a new generation of high-performance photoelectric detection technology.

[0003] A two-dimensional material linear array detection imaging system is an imaging device constructed based on a two-dimensional material linear array detector, which can obtain the spatial light intensity distribution of a target object through a mechanical scanning method, and then achieve high-resolution image acquisition. The working process of this system is that first, the linear array detector performs photoelectric conversion on the incident light and outputs a response signal; then the signal is processed by an amplifier circuit and then undergoes analog-to-digital conversion; finally, the digital signal generates an image after subsequent processing to achieve the target imaging function.

[0004] Compared with the linear array detectors made of traditional bulk materials, although the two-dimensional material linear array detectors have relatively high response speed and sensitivity at the single pixel level, at present, two-dimensional materials such as graphene mainly rely on methods such as mechanical exfoliation or chemical vapor deposition for preparation. In addition, there are still technical bottlenecks in the precision control and interface stability of the array transfer process, resulting in poor non-uniformity of the overall device. In addition, due to factors such as the manufacturing precision differences, process inconsistencies, and uneven optical path distributions of the detector's backend circuits, additional non-uniformity errors will be introduced into the system. Therefore, the directly acquired images often show bright and dark alternating stripes along the scanning direction of the detector, forming fixed pattern noise. Such noise not only significantly reduces the signal-to-noise ratio of the image, resulting in the loss of detail information, but also affects the recognition and measurement accuracy of the target, thus severely restricting the improvement of imaging quality. In addition, due to the generally high dark current in two-dimensional material detectors themselves, if not suppressed, it will further reduce the contrast of the image and affect the system performance.

[0005] At present, non-uniformity correction mainly relies on reference source calibration methods based on software algorithms, including single-point calibration, two-point calibration, and multi-point calibration, etc. Such methods can alleviate the problem of inconsistent responses among detectors to a certain extent. However, for two-dimensional material linear array detectors, due to their particularly prominent non-uniformity problems, there are often individual detection units in the system with extremely deviated response performances. These abnormal pixels will significantly interfere with the overall average value during the calibration process, thereby affecting the accurate estimation of the gain coefficient and bias coefficient, weakening the correction effect of traditional calibration algorithms, and even possibly causing the calibration of some detectors to fail. Summary of the Invention

[0006] Technical Problem: The object of the present invention is to address the non-uniformity problem existing in two-dimensional material linear array detection imaging systems, and propose a non-uniformity correction method based on two-dimensional material linear array detection imaging systems. This method adjusts the gain and bias of each detector at the hardware level to achieve preliminary correction of the consistency of optoelectronic responses; subsequently, a calibration algorithm is introduced at the software level to finely compensate for the residual non-uniformity error, thereby effectively improving the consistency of system imaging and image quality, and significantly improving the correction effect of traditional methods in practical applications.

[0007] Technical Solution: The non-uniformity correction method for a two-dimensional material linear array detector imaging system of the present invention includes the following steps:

[0008] Step S1. Adjust the voltage amplification factor of each channel according to the photocurrent; measure the light and dark currents of each detector in the linear array detector, and determine the amplified voltage of the optical signal based on the ratio of the minimum light and dark currents. Use devices such as digital potentiometers as feedback resistors in the amplification circuit and other methods to adjust the amplification factors of each path to make the output voltages of each path of optical signals consistent;

[0009] Step S2. Eliminate the amplified dark current in a differential manner according to the amplification factor; use a dedicated differential amplification chip or an analog-to-digital conversion chip that can achieve differential input conversion to implement differential amplification and remove the voltage amplified by the dark current;

[0010] Step S3. Obtain two-point calibration parameters through line scanning; perform multi-column mechanical scanning on the linear array detector at different powers to obtain richer data samples, and calculate the gain coefficient and bias coefficient of two-point correction;

[0011] Step S4. Correct the output of the detector with the correction parameters of two-point calibration. Correct the original output voltage of each detector with the gain coefficient and bias coefficient of two-point correction.

[0012] Among them,

[0013] The specific content of step S1 is as follows:

[0014] Step S1.1. Measure the dark current and total current of each detector in the linear array detector to obtain the dark current I of M detectors dark and the total current I total ,

[0015] Step S1.2. Calculate the photocurrent I photo , I photo = I total - I dark ; Calculate the ratio LDR of the photocurrent to the dark current of each detector,

[0016] Step S1.3. Select the detector with the lowest LDR and the total current I total Determine the amplification factor A of this path LDR,min , and under the constraints of various conditions, increase the amplification factor as much as possible to enhance the imaging contrast at the backend;

[0017] Step S1.4. Taking the amplification voltage V photo of the detector with the lowest LDR as a reference, adjust the amplification factor A of each other detector so that the voltage value after amplification of the photocurrent of other detectors is equal to the amplification voltage of Ix photo of the detector with the lowest LDR. hoto of the detector with the lowest LDR.

[0018] The specific content of S2 is as follows:

[0019] Step S2.1. Calculate the amplification voltage V dark (m) of the dark current of each detector dark , V dark (m)= I dark (m)×A(m);

[0020] Step S2.2. Use the differential method to filter out the voltage V total (m) amplified by the dark current in the amplified voltage V dark (m) of the detector to obtain the voltage V photo (m) amplified by the photocurrent, V photo (m)= V total (m)- V dark (m).

[0021] The specific content of S3 is as follows:

[0022] Step S3.1. Perform a linear scan of N columns on the linear array detector under illumination with powers p1 and p2 to obtain the M×N data matrices V m,n (p1) and V m,n (p2), where m represents the m-th detector of the linear array detector and n represents the n-th column of data;

[0023] Step S3.2. Calculate the average value of the m-th detector at power p1

[0024] Step S3.3. Calculate the average value of the m-th detector at power p2

[0025] Step S3.4. Calculate the average value of the linear array detector at power p1

[0026] Step S3.5. Calculate the average value of the linear array detector at power p2

[0027] Step S3.6. Let the corrected gain coefficient be P m , and the bias coefficient be Q m , According to the two-point calibration theorem

[0028] Derive the expressions of the gain coefficient P m and the bias coefficient Q m

[0029]

[0030] The specific content of S4 is as follows: Use the gain coefficient P m , and the bias coefficient Q m to correct the original output value V m (p) of the detector into the value V' for imaging m , where V' m =P m ×V m (p)+Q m .

[0031] Advantageous effects: The above technical solution of the present invention has the following advantages compared with the prior art:

[0032] This software and hardware collaborative non-uniformity correction method can effectively overcome the problem that the correction effect of the traditional software algorithm-based reference source calibration method is seriously affected or even fails when dealing with detection units with extremely deviated response performance. This method can not only effectively filter out the large dark current of each detection pixel in the linear array detector of two-dimensional materials, but also solve the problem of the difference in optical response between each detector by adjusting the amplification factor of each channel, realizing the consistency of the output of the optical response signal. The preliminary non-uniformity correction at the hardware level combined with the linear array scanning imaging obtains more data samples, providing a more stable data basis for the subsequent software two-point calibration to correct non-uniformity, making the calculated gain coefficient and bias coefficient more accurate, thus significantly improving the overall correction effect and image quality.​ Detailed implementation mode

[0033] The non-uniformity correction method for the two-dimensional material linear array detector imaging system of the present invention includes the following steps:

[0034] Step S1. Adjust the voltage amplification factor of each channel according to the photocurrent; measure the light and dark currents of each detector in the linear array detector, and determine the amplified voltage of the optical signal based on the ratio of the minimum light and dark currents. Use devices such as digital potentiometers as the feedback resistors in the amplification circuit to adjust the amplification factors of each channel so that the output voltages of each optical signal are consistent;

[0035] Step S2. Eliminate the amplified dark current in a differential manner according to the amplification factor; use a dedicated differential amplification chip or an analog-to-digital conversion chip that can implement differential input conversion to achieve differential amplification and remove the voltage amplified by the dark current;

[0036] Step S3. Obtain two-point calibration parameters through line scanning; perform multi-column mechanical scanning on the linear array detector at different powers to obtain richer data samples, and calculate the gain coefficient and bias coefficient for two-point correction;

[0037] Step S4. Correct the output of the detector with the two-point calibration correction parameters. Use the gain coefficient and bias coefficient for two-point correction to correct the original output voltage of each detector.

[0038] Specifically:

[0039] Step S1

[0040] Measure the dark current I of each detector in the 1×8 linear array detector dark and the total current I total , and obtain the dark currents of 8 detectors, which are 593 nA, 357 nA, 317 nA, 289 nA, 401 nA, 533 nA, 354 nA, and 469 nA respectively. The corresponding total currents are 1.377 μA, 889 nA, 932 nA, 740 nA, 1.39 μA, 1.43 μA, 757 nA, and 1.175 μA respectively. Through the formula I photo =I total -I dark calculation, the photocurrents corresponding to each detector can be obtained as 784 nA, 532 nA, 615 nA, 451 nA, 989 nA, 897 nA, 403 nA, and 706 nA respectively;

[0041] According to the formula calculate the ratio LDR of the photocurrent to the dark current of each detector, which are 1.322, 1.490, 1.940, 1.561, 2.466, 1.683, 1.138, and 1.505 respectively;

[0042] The smallest LDR is 1.138. Taking this detector as a reference and comprehensively considering various requirements in the amplifier circuit, the output voltage value of one stage of the amplifier circuit cannot exceed 2.5V. Therefore, the voltage value after amplification of this stage for the detector with an LDR of 1.138 cannot exceed 2.5V. After amplifying the total current of this detector, the obtained voltage value is 2.35V, and the voltage value of the photocurrent amplification is 1.25V;

[0043] Taking 1.25V as the voltage value converted and output by the photocurrent of all detectors in this stage of the amplifier circuit, according to the photocurrent of other detectors and the requirement that the photocurrent should be amplified and output as 1.25V, the amplification factor A(m) of the photocurrent of the m-th detector can be determined. By using a digital potentiometer as the feedback resistor in the amplifier circuit and other means, the amplification factor of the m-th detector can be freely adjusted to A(m);

[0044] Step S2

[0045] According to the formula V dark (m) = I dark (m) × A(m);, the voltage value V dark (m) of the dark current amplification can be calculated, which are 0.945V, 0.839V, 0.644V, 0.801V, 0.507V, 0.743V, 1.098V, and 0.830V respectively. An 8-channel output digital-to-analog conversion chip can be used to output each voltage value of the dark current amplification. A dedicated differential amplification chip or an analog-to-digital conversion chip that can implement differential input is used to total make a difference with the total current amplification voltage V

[0046] Step S3

[0047] The linear array detector is linearly scanned for 8 columns under illuminations with powers p1 and p2, obtaining 8×8 data matrices V m,n (p1) and V m,n (p2), where m represents the m-th detector in the linear array detector and n represents the n-th column data in the 8 columns;

[0048] Calculate the average value of the m-th detector at power p1

[0049] Calculate the average value of the m-th detector at power p2

[0050] Calculate the average value of the linear array detector at power p1

[0051] Calculate the average value of the linear array detector at power p2

[0052] Let the corrected gain coefficient be P m , and the bias coefficient be Q m , according to the two-point calibration theorem

[0053] Derive the gain coefficient P m and the bias coefficient Q m expression

[0054]

[0055] Step S4

[0056] After two-point correction, the m-th detector can obtain the gain coefficient P m and the bias coefficient Q m , use P m and Q m to correct the original output voltage value V m (p) of the detector, and obtain the corrected voltage value V' m , V' m = P m ×V m (p) + Q m , use V' m for subsequent imaging, which realizes non-uniformity correction at the software level;

[0057] Experimental results:

[0058] By scanning and imaging the same object, and comparing the original image, the image corrected by the traditional two-point calibration method, and the imaging results of the software and hardware collaborative non-uniformity correction method proposed in the present invention, the software and hardware collaborative method has a better effect in reducing the severity of the light and dark stripes in the image, and at the same time significantly improves the overall contrast of the image. This shows that the method can more effectively suppress the influence of non-uniformity in all aspects of the system, thereby improving the imaging quality and system stability.

Claims

1. A non-uniformity correction method for a two-dimensional material line array detection imaging system, characterized in that The method includes: Step S1. Adjust the voltage amplification factor of each channel according to the photocurrent; measure the light and dark currents of each detector in the linear array detector, and determine the amplified voltage of the optical signal based on the ratio of the minimum light and dark currents. Use devices such as digital potentiometers as the feedback resistors in the amplification circuit to adjust the amplification factors of each path, so that the output voltages of each path of optical signals are consistent; Step S2. Eliminate the amplified dark current in a differential manner according to the amplification factor; use a dedicated differential amplification chip or an analog-to-digital conversion chip that can implement differential input conversion to achieve differential amplification and remove the voltage amplified by the dark current; Step S3. Obtain two-point calibration parameters through linear array scanning; perform multi-column mechanical scanning on the linear array detector at different powers to obtain richer data samples, and calculate the gain coefficient and bias coefficient of two-point correction; Step S4. Correct the output of the detector with the correction parameters of two-point calibration. Correct the original output voltage of each detector using the gain coefficient and bias coefficient of two-point correction.

2. The non-uniformity correction method for a two-dimensional material line array detection imaging system according to claim 1, wherein Specifically, step S1 is as follows: Step S1.

1. Measure the dark current and total current of each detector in the linear array detector to obtain the dark current I of M detectors dark and the total current I total , Step S1.

2. Calculate the photocurrents I of M detectors photo , I photo = I total - I dark ; Calculate the ratio LDR of the photocurrent to the dark current for each detector, Step S1.

3. Select the total detector current I with the lowest LDR total Determine the amplification factor A of this path LDR,min , and under the constraints of various conditions, increase the amplification factor as much as possible to enhance the imaging contrast at the backend; Step S1.

4. Using the amplified voltage V of the detector I with the lowest LDR photo as a reference, adjust the amplification factor A of each of the other detectors so that the voltage value after the photocurrent of the other detectors is amplified is equal to the amplified voltage of the detector I with the lowest LDR photo . photo ​ 3. The non-uniformity correction method for a two-dimensional material line array detection imaging system according to claim 2, wherein Specifically, S2 is as follows: Step S2.

1. Calculate the amplified voltage V dark of each detector dark current I dark (m), V dark (m) = I dark (m) × A(m); Step S2.

2. Using the differential method, filter out the voltage V total (m) of the amplified dark current in V dark (m) to obtain the voltage V photo (m) of the amplified photocurrent, V photo (m) = V total (m) - V dark (m).

4. The non-uniformity correction method of the two-dimensional material line array detection imaging system according to claim 3, characterized in that, Specifically, S3 is as follows: Step S3.

1. Perform line scanning on the line array detector for N columns under illuminations with powers p1 and p2 to obtain an M×N data matrix V m,n (p1) and V m,n (p2), where m represents the m-th detector of the line array detector and n represents the n-th column of data; Step S3.

2. Calculate the average value of the m-th detector at power p1 Step S3.

3. Calculate the average value of the m-th detector at power p2 Step S3.

4. Calculate the average value of the linear array detector at power p1 Step S3.

5. Calculate the average value of the linear array detector at power p2 Step S3.

6. Set the calibrated gain coefficient as P m , and the bias coefficient as Q m , according to the two-point calibration theorem Derive the expression for the gain coefficient P m and the bias coefficient Q m of the expression 5. The non-uniformity correction method for a two-dimensional material line array detection imaging system according to claim 4, wherein, Specifically, S4 is as follows: Use the gain coefficient P m , the bias coefficient Q m to correct the original output value V m (p) of the detector into the value V' for imaging m , where V' m = P m ×V m (p) + Q m .

Images