Method for improving line frequency matching accuracy of multispectral TDI detector

Abstract

The present application relates to multispectral TDI detector technical field, especially to a kind of multispectral TDI detector line frequency matching precision promotion method, multispectral TDI detector includes at least two spectral bands, each spectral band has the row period of respective length, and the drive control signal sent to multispectral TDI detector by imaging controller only exists one timing reset signal, and the cycle period of this timing reset signal uses the longest row period in each spectral band of multispectral TDI detector to reset each spectral band counter in imaging controller internal;The main clock frequency of the timing signal of each spectral band in imaging controller internal is the integer multiple of pixel clock frequency;Charge transfer timing signal is adjusted according to the rule of equal multiple adjustment;Analog-digital conversion and subsequent control signal then when the length of multiple sub row period reaches the integer multiple of pixel clock, the timing position of adjustment is carried out, to achieve the matching precision of sub-pixel clock cycle length under single timing reset signal state.

Description

Technical Field

[0001] This invention belongs to the field of multi-band TDI detector technology, and particularly relates to a method for improving the line frequency matching accuracy of multi-band TDI detectors. Background Technology

[0002] Time-delay integration (TDI) detectors can increase the received light energy by hundreds of times by accumulating and integrating the same scene, reducing the need for optical engine aperture and thus reducing the size and weight of the optical engine. In TDI imaging, improving the matching degree between charge transfer within the detector and image movement can further improve the modulation transfer function of the image during dynamic imaging. For a TDI detector, in addition to the photoelectric conversion section and charge drive control section, it also includes digital quantization of the detector's output analog signal. The relevant operating timing references the pixel clock of the TDI detector; that is, the relevant operating timing positions are all based on the pixel clock length as the smallest step size. The line period length of the TDI detector is an integer multiple of the pixel period length. The line period matching accuracy is limited by the resolution of the pixel clock period; therefore, improving the line period resolution is of great significance for fine image movement matching.

[0003] The patent with publication number CN102123254A and publication date July 25, 2012 discloses a timing control method to reduce image shift in multi-phase TDICCD. This method proposes to divide the charge transfer process of one cycle into 2n equally spaced time periods, where n is the number of phases of charge transfer. Theoretically, this can achieve the maximum dynamic transfer function, but it ignores the mutual interference between spectral bands. At the same time, it ignores the bottleneck of the maximum full-well that the detector can obtain, which is the maximum full-well during the charge transfer process. Using a single electrode to control the charge transfer will reduce the available maximum full-well problem and affect the maximum signal-to-noise ratio and temporal uniformity of the image.

[0004] The patent with publication number CN108401105A and publication date August 21, 2020 discloses a method for improving the dynamic transfer function of a space remote sensing TDICCD camera and a space camera. This method is an improvement on CN102123254A. While ensuring that the charge transfer dynamic modulation transfer function is 0.9432, the full-well charge is comparable to the non-uniform charge transfer timing mode. However, it ignores the mutual interference between spectral bands, which will affect the temporal uniformity of the image.

[0005] Patent CN113364933A, published on May 10, 2022, discloses a TDICMOS imaging system with high camera synchronization and fine image shift compensation. It addresses the issue of abrupt changes in the driving signal edge during line period length variations, which can lead to charge transfer errors and abnormal images. The system uses the sum of four adjacent full-color line periods instead of quadrupling them for the multispectral line period. It also establishes a protection zone at the update position of the charge transfer signal and modifies the output judgment condition for the charge transfer signal, specifically judging the phase relationship between the rising and falling edges and using the signal level duration instead of the signal edge judgment. This allows it to operate under arbitrary line period step changes. However, this patent does not address the root cause of the problem in timing design; it only mitigates the adverse effects at the back end, requiring significant logic resources.

[0006] Patent CN111510647A, published on August 31, 2021, discloses a uniform charge transfer control method for multi-band TDICMOS. This method proposes setting the minimum pixel clock count corresponding to the minimum row period of the detector's full color to an integer power of 2. This eliminates the need for division when calculating the row period length in real-time, requiring only multiplication followed by truncating the high-order bits. However, this patent also has drawbacks. It fails to avoid negative values ​​during the calculation of new transition edges when the row period changes, thus avoiding the omission of calculation steps and saving computational resources. Furthermore, it fails to prevent timing errors caused by transition edges.

[0007] Patent CN117768798A, published on March 26, 2024, discloses a charge transfer method for multi-band TDICMOS. This method proposes that vertical charge transfer is not completely uniformly divided, but rather, after deducting sensitive areas that interfere with the image, the remaining area is transferred as uniformly as possible. The maximum full-well size achievable during charge transfer is n times the maximum full-well size achieved using a single electrode to control the charge. The complexity of the implementation process is reduced by optimizing the starting position of the timing design and the transition edge setting of the drive control signal. However, this patent is based on the line period adjustment step size being an integer multiple of the pixel clock, and does not consider the implementation problem of setting the line period adjustment precision to sub-pixel clock precision.

[0008] Patent CN116709044A, published on September 5, 2023, discloses a method for improving the line frequency matching accuracy of a TDI detector. This method proposes that each drive control signal, under the synchronous control of the serial image data clock, is generated based on the count values ​​of the first and second stage counters. A new timing start point for a line can only be reset and re-counting can only begin when the count values ​​of the first and second stage counters in the previous line have both reached their set values. This patent is applicable to single-spectral-band applications or applications where all spectral-band lines have the same line period. For multi-spectral-band TDI detector applications, such as when the line period length of the multispectral band is an integer multiple of the panchromatic line period length, independent timing reset signals are required for different line period lengths. Therefore, this method cannot be applied to detectors with only one timing reset signal. Summary of the Invention

[0009] In view of this, the present invention aims to provide a method for improving the line frequency matching accuracy of a multi-band TDI detector, so as to solve the problem of achieving line frequency matching of a multi-band TDI detector with sub-pixel clock cycle accuracy under a single timing reset signal.

[0010] To achieve the above objectives, the technical solution created by this invention is implemented as follows: A method for improving the line frequency matching accuracy of a multi-band TDI detector is disclosed. The multi-band TDI detector includes at least two spectral bands, each with its own line period of length. The driving control signal sent by the imaging controller to the multi-band TDI detector contains only one timing reset signal. The cycle period of this timing reset signal is based on the longest line period among the spectral bands of the multi-band TDI detector, which resets the counters of each spectral band inside the imaging controller. The master clock frequency of the timing signal of each spectral band inside the imaging controller is an integer multiple of the pixel clock frequency.

[0011] Furthermore, the timing signals of each spectral band within the imaging controller include charge transfer timing signals, and the position of the transition edge of the charge transfer timing signal is determined according to the length of the input line period.

[0012] Furthermore, the timing signals for each spectral band within the imaging controller also include analog-to-digital conversion and subsequent control signals; When the line cycle length input to the analog-to-digital converter and subsequent control signals is an integer multiple of the pixel clock cycle, the following two cases apply: For spectral bands in a multi-band TDI detector with pixel sizes smaller than a preset value, add the difference between the current line period length and the default line period length in the multi-band TDI detector manual to the end of the line period of each spectral band. For spectral bands in a multi-band TDI detector with pixel sizes larger than a preset value, the row period of each spectral band is evenly divided into equally divided sub-row periods, and the difference between the current row period length and the default row period length in the multi-band TDI detector manual is added to the end of each sub-row period. When the row cycle length input to the analog-to-digital converter and subsequent control signals is a non-integer multiple of the pixel clock cycle, there are two cases: the integer part and the non-integer part. Regarding the integer part: For spectral bands in a multi-band TDI detector with pixel sizes smaller than a preset value, add the difference between the current line period length and the default line period length in the multi-band TDI detector manual to the end of the line period of each spectral band. For spectral bands in a multi-band TDI detector with pixel sizes larger than a preset value, the row period of each spectral band is evenly divided into equally divided sub-row periods, and the difference between the current row period length and the default row period length in the multi-band TDI detector manual is added to the end of each sub-row period. Regarding the non-integer part: The ratio of the non-integer part to the pixel clock cycle length is β, and it satisfies βε=k, that is, the non-integer part becomes an integer multiple of the pixel clock k after ε cycles of accumulation, where k is a prime number; For spectral bands in a multi-band TDI detector where the pixel size is smaller than a preset value, add k pixel clock cycles at the end of every ε row period; For spectral bands in a multi-band TDI detector with pixel sizes larger than a preset value, the row period of each spectral band is evenly divided into equally divided sub-row periods, and k additional pixel clock periods are added at the end of each ε sub-row period.

[0013] Furthermore, in spectral segments with different line period lengths, the ratio of the line period lengths of any two spectral segments is a positive integer.

[0014] Compared with the prior art, the present invention can achieve the following beneficial effects: 1. The timing signal for charge transfer is an analog control signal with no concept of time resolution. Therefore, even if the input row period is not an integer multiple of the pixel clock, it is still performed in the manner of uniform charge transfer, thereby obtaining the matching accuracy of the sub-pixel clock period. As for the control signals of analog-to-digital conversion and subsequent stages, they belong to the control signals of the internal digital circuit of the detector and are controlled by the pixel clock. Therefore, it is necessary to round up the sub-pixel clock periods of multiple fractional parts to make them integer multiples of the pixel clock period before control, so as to ensure the stability and reliability of the operation of the internal digital circuit of the detector.

[0015] 2. Among the drive control signals sent by the imaging controller to the detector, there is only one timing reset signal, which can reduce the complexity of the detector's internal design and ensure that each spectral band generates a timing start point under the control of the same control signal.

[0016] 3. The master clock frequency of the timing signals of each spectral band inside the imaging controller is an integer multiple of the pixel clock frequency, which facilitates fine control of the line period. At the same time, all control signals work under the same clock cycle, ensuring the reliability and stability of the timing. Attached Figure Description

[0017] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments and descriptions of the invention are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a schematic diagram of the logical structure of the imaging system of the multi-band TDI detector according to an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not constitute a limitation thereof.

[0019] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0020] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0021] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0022] The following will refer to Figure 1 The invention will be described in detail with reference to the embodiments.

[0023] This invention provides a method for improving the line frequency matching accuracy of a multi-band TDI detector, which is implemented through the imaging system of the multi-band TDI detector. For example... Figure 1 As shown, the imaging system of the multi-band TDI detector includes a camera controller, an imaging controller power supply chip, an imaging controller, a detector power supply chip, a multi-band TDI detector, erasable main and backup flash memory, a PROM, an MRAM, a CameraLink chip, a CameraLink connector, a 2711 chip, and a 2711 connector. The camera controller receives a primary power supply from an external input and generates the necessary power supplies; it also receives a second pulse from an external input and communicates with the external system via a 1553 bus. The camera controller communicates with the imaging controller via a 422 communication signal and provides the second pulse to the imaging controller, supplying power to the imaging controller through the imaging controller power supply chip. The imaging controller provides drive control signals (with only one timing reset signal) to the multi-band TDI detector and supplies power to the multi-band TDI detector through the detector power supply chip, receiving serial image data output from the multi-band TDI detector. The imaging controller is simultaneously connected to the configuration data source PROM and the erasable main and backup flash memory, and also connected to the MRAM for updating correction coefficients and loading correction coefficients before each image capture. The imaging controller can output image data via the Cameralink chip and Cameralink connector, or via the 2711 chip and 2711 connector, depending on the selection.

[0024] The imaging controller power supply chip uses the 510 DC-DC module; the detector uses a field array detector from Changguang Chenxin Company; the camera controller mainly uses a DSP chip; the imaging controller mainly uses the imaging controller and refresh chip from Shanghai Fudan Microelectronics Company; the detector power supply chip mainly uses an LDO from TI Company; the 2711 chip uses the TLK2711 chip; the 2711 connector uses a micro coaxial connector from Sichuan Huafeng; the Cameralink chip uses the DS90CR287; the Cameralink connector uses the MDR26 connector from 3M Company; the erasable main and backup flash memory uses products from Shanghai Fudan Microelectronics Company; the PROM uses products from Xilinx Company; and the MRAM uses products from the 771 Institute.

[0025] A multi-band TDI detector includes at least two spectral bands, each with its own line period of length. Some spectral bands have the same line period length, while others have different line period lengths. For spectral bands with different line period lengths, the ratio of the line period lengths of any two of them is a positive integer.

[0026] A method for improving the line frequency matching accuracy of a multi-band TDI detector includes: the driving control signal sent by the imaging controller to the multi-band TDI detector contains only one timing reset signal, the cycle period of which adopts the longest line period among the various spectral bands of the multi-band TDI detector, in order to reset the counters of each spectral band inside the imaging controller.

[0027] The longest line period among the multi-spectral bands is chosen as the cycle period of the timing reset signal. This ensures that the counters of each spectral band within the imaging controller are synchronously reset at the end of the longest line period, guaranteeing complete alignment of the timing start points for each spectral band within the imaging controller. For example, when the longest line period ends, all spectral band counters return to zero, starting a new cycle. This ensures that even if the line periods of each spectral band are different, their timing start points are the same, preventing misalignment. By sharing a single timing reset signal, the timing control start point of the multi-spectral TDI detector is unified across all spectral bands. The multi-spectral TDI detector does not require separate reset logic for each spectral band; it only needs to process a single timing reset signal. This reduces logic gates and wiring, simplifying the internal logic of the multi-spectral TDI detector and thus lowering the complexity of its internal circuit design. Furthermore, the unified timing start point facilitates data processing and synchronization across spectral bands, avoiding errors caused by asynchrony.

[0028] The master clock frequency of the timing signals for each spectral band inside the imaging controller is an integer multiple of the pixel clock frequency, which facilitates fine control of the line period. At the same time, all control signals operate at the same beat, ensuring the reliability and stability of the timing.

[0029] In addition to the timing reset signal shared by all spectral bands, the timing signals of each spectral band inside the imaging controller are divided into charge transfer timing signals and analog-to-digital conversion and subsequent control signals. The processing methods of charge transfer timing signals and analog-to-digital conversion and subsequent control signals are different. The imaging controller achieves precise matching of line periods by adjusting the transition edge positions of the timing signals (such as charge transfer signals and analog-to-digital conversion signals) of each spectral band.

[0030] For adjusting the charge transfer timing signal, the transition edge positions of each signal are adjusted according to the input line period value, following the charge transfer method in patent CN117768798A. Since the charge transfer timing signal is an analog control signal and has no concept of time resolution, it will still be processed in a uniform charge transfer manner even if the input line period is not an integer multiple of the pixel clock, thereby achieving sub-pixel clock period matching accuracy.

[0031] For the adjustment of analog-to-digital conversion and subsequent control signals, there are two cases: one where the input line period length is an integer multiple of the pixel clock period, and the other where the input line period length is a non-integer multiple of the pixel clock period. The adjustment methods for these two cases differ. In other words, the timing adjustment of analog-to-digital conversion and subsequent control signals needs to consider the integer multiple relationship of the pixel clock. If the input line period length is an integer multiple of the pixel clock period, the line period length is directly adjusted; if the input line period length is not an integer multiple of the pixel clock period, the non-integer portion of the pixel clock period is accumulated until an integer multiple of the pixel clock period is achieved.

[0032] More specifically, when the length of the input row period is an integer multiple of the pixel clock period, the following two cases apply depending on the pixel size: ① For spectral bands in a multi-band TDI detector where the pixel size is smaller than the preset value, add the difference between the current line period length and the default line period length in the multi-band TDI detector manual to the end of the line period of each spectral band.

[0033] ② For spectral bands in a multi-band TDI detector where the pixel size is larger than a preset value, the row period of each spectral band is evenly divided into equal sub-row periods, and the difference between the current row period length and the default row period length in the multi-band TDI detector manual is added to the end of each sub-row period.

[0034] When the length of the input row period is a non-integer multiple of the pixel clock period, there are two cases: the integer part and the non-integer part.

[0035] ① Integer part: For spectral bands in a multi-band TDI detector with pixel sizes smaller than a preset value, add the difference between the current line period length and the default line period length in the multi-band TDI detector manual to the end of the line period of each spectral band.

[0036] For spectral bands in a multi-band TDI detector with pixel sizes larger than a preset value, the row period of each spectral band is evenly divided into equally divided sub-row periods, and the difference between the current row period length and the default row period length in the multi-band TDI detector manual is added to the end of each sub-row period.

[0037] ② Non-integer part: The ratio of the non-integer part to the pixel clock cycle length is β, and it satisfies βε=k, that is, the non-integer part accumulates for ε cycles and becomes an integer multiple of the pixel clock k, where k is a prime number.

[0038] For spectral bands in a multi-band TDI detector with pixel sizes smaller than a preset value, k pixel clock cycles are added at the end of every ε row cycles (based on an increase of an integer multiple of the pixel clock row cycle).

[0039] For spectral bands in a multi-band TDI detector with pixel sizes larger than a preset value, the row period of each spectral band is evenly divided into equally divided sub-row periods, and k additional pixel clock periods are added at the end of each ε sub-row period (based on an increase of an integer multiple of the pixel clock row period).

[0040] The line period length of each spectral segment is an integer multiple of the pixel clock period to ensure coordination between different spectral segments.

[0041] For non-integer multiples of the row period length, the non-integer portion of the pixel clock period is accumulated to eventually achieve an integer multiple of the pixel clock period, thus avoiding timing jump issues.

[0042] The analog-to-digital conversion and subsequent control signals belong to the control signals of the internal digital circuit of the multi-band TDI detector. They are controlled by the pixel clock. Therefore, it is necessary to round up the sub-pixel clock cycles of multiple fractional parts to make an integer multiple of the pixel clock cycle before control, so as to ensure the stability and reliability of the internal digital circuit of the detector.

[0043] This invention allows the line period adjustment step size to be sub-pixel clock cycles (non-integer multiples). By uniformly adjusting the charge transfer timing signal, performing analog-to-digital conversion, and compensating for integer accumulation in subsequent control signals, it breaks through the traditional integer multiple limitation, significantly improving image shift matching accuracy. Through precise image shift matching and high-resolution line period control, it effectively improves the dynamic modulation transfer function (MTF), reduces image blur, enhances the clarity and signal-to-noise ratio of remote sensing imaging, and significantly improves the imaging quality of multi-band TDI detectors.

[0044] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this invention disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.

[0045] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for improving the line frequency matching accuracy of a multi-band TDI detector, wherein the multi-band TDI detector comprises at least two spectral bands, each spectral band having its own line period of length, characterized in that, include: The driving control signal sent by the imaging controller to the multi-band TDI detector contains only one timing reset signal. The cycle period of this timing reset signal is based on the longest line period among the various spectral bands of the multi-band TDI detector to reset the counters of each spectral band inside the imaging controller. The master clock frequency of the timing signal of each spectral band inside the imaging controller is an integer multiple of the pixel clock frequency. The timing signal of each spectral band inside the imaging controller includes the charge transfer timing signal, and the position of the transition edge of the charge transfer timing signal is determined according to the length of the input line period. The timing signals for each spectral band within the imaging controller also include analog-to-digital conversion and subsequent control signals; When the line cycle length input to the analog-to-digital converter and subsequent control signals is an integer multiple of the pixel clock cycle, the following two cases apply: For spectral bands in a multi-band TDI detector with pixel sizes smaller than a preset value, add the difference between the current line period length and the default line period length in the multi-band TDI detector manual to the end of the line period of each spectral band. For spectral bands in a multi-band TDI detector with pixel sizes larger than a preset value, the row period of each spectral band is evenly divided into equally divided sub-row periods, and the difference between the current row period length and the default row period length in the multi-band TDI detector manual is added to the end of each sub-row period. When the row cycle length input to the analog-to-digital converter and subsequent control signals is a non-integer multiple of the pixel clock cycle, there are two cases: the integer part and the non-integer part. Regarding the integer part: For spectral bands in a multi-band TDI detector with pixel sizes smaller than a preset value, add the difference between the current line period length and the default line period length in the multi-band TDI detector manual to the end of the line period of each spectral band. For spectral bands in a multi-band TDI detector with pixel sizes larger than a preset value, the row period of each spectral band is evenly divided into equally divided sub-row periods, and the difference between the current row period length and the default row period length in the multi-band TDI detector manual is added to the end of each sub-row period. Regarding the non-integer part: The ratio of the non-integer part to the pixel clock cycle length is β, and it satisfies βε=k, that is, the non-integer part becomes an integer multiple of the pixel clock k after ε cycles of accumulation, where k is a prime number; For spectral bands in a multi-band TDI detector where the pixel size is smaller than a preset value, add k pixel clock cycles at the end of every ε row period; For spectral bands in a multi-band TDI detector with pixel sizes larger than a preset value, the row period of each spectral band is evenly divided into equally divided sub-row periods, and k additional pixel clock periods are added at the end of each ε sub-row period.

2. The method for improving the line frequency matching accuracy of a multi-band TDI detector according to claim 1, characterized in that, In spectral segments with different line period lengths, the ratio of the line period lengths of any two segments is a positive integer.

Images