A method of fine training of a probe

Abstract

The present application relates to a kind of fine training method of detector, for the application of well-controlled temperature environment and the application of badly-controlled temperature environment, according to the number of stable region and the number of unstable region sampled are classified and are respectively delayed to the delay of serial image data and sampling clock, and sampling region is expanded, complete sampling information is obtained, so as to set the best sampling position.The present application is directed to only using serial image data to delay sampling to an unstable region, it cannot be determined that the nearest unstable region of adjacent distance appears on the left side or right side, it is proposed to delay the sampling clock on the basis of serial data delay, so as to obtain more complete unstable region distance information, so as to set the best sampling position, improve temperature adaptability.

Description

Technical Field

[0001] This invention relates to the field of detector technology, and in particular to a method for fine training of a detector. Background Technology

[0002] During detector training, traditional methods generally work without issues. However, defects in the detector's internal wiring or circuit board layout can lead to an additional unstable region in the serial data being detected during bit correction, beyond the normal two unstable regions. Furthermore, in addition to the three typical states (no edge detected, one edge detected, and two edges detected), an extra edge position may appear. This means it might detect two stable length regions with three edges, two edges with one longer or shorter stable region, or only one edge position. For applications where the serial image data frequency is less than half the reference clock frequency, dual-edge sampling of the sampling clock can be used to expand the sampling area. However, for applications where the serial image data frequency is greater than half the reference clock frequency, using dual-edge sampling of the sampling clock to expand the sampling area may lead to misjudgment of sampling positions due to overlapping sampling areas. Without bit width expansion, incomplete sampling states may result in poor environmental adaptability. Summary of the Invention

[0003] The present invention aims to solve the technical problems in the prior art by providing a method for fine training of a detector.

[0004] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0005] A fine-tuning method for a detector, suitable for applications with well-controlled temperature environments, includes the following steps:

[0006] When three unstable regions and two stable regions are detected, the center of the region with the longer stable region is selected as the final sampling location.

[0007] When a long stable region and two unstable regions are detected, and the sum of the widths of the long stable region and the unstable region is greater than half of the total number of delay taps, the sampling position is the midpoint of the long stable region.

[0008] When a short stable region and two unstable regions are detected, characterized by the sum of the widths of the long stable region and one unstable region being less than or equal to half of the total number of delay taps, the sampling location is:

[0009] Maximum number of taps + left edge of the first unstable region - (maximum number of taps - width of two unstable regions - width of the short stable region) / 2 =

[0010] The left edge of the first unstable region + half of the maximum number of taps - (width of the two unstable regions + width of the short stable region) / 2;

[0011] When only one unstable region is detected:

[0012] When the unstable region is at the exact center, the sampling position is set to half of the left edge of the unstable region;

[0013] When the unstable region is on the left, the sampling position is set to the right edge of the unstable region + (maximum number of taps - right edge of the unstable region) / 2 = half of the maximum number of taps + half of the right edge of the unstable region;

[0014] When the unstable region is on the right, the sampling position is set to half of the left edge of the unstable region;

[0015] For applications with harsh temperature control conditions, the following steps are included:

[0016] When the serial image data frequency is less than or equal to half of the reference clock frequency, the sampling area is widened by dual-edge sampling based on the sampling clock.

[0017] When the serial image data frequency is greater than 0.5 times the reference clock frequency but less than r times the reference clock frequency, 0.5 < r < 0.6, the length range of the detected stable region is limited. At the same time, when a transition edge is detected, the distance between the sampling position and the transition edge is limited. The widened sampling region is obtained based on the dual-edge sampling of the sampling clock.

[0018] After clock delay extension, if the stable region on the right is long, the stable sampling point is the right side of the unstable region plus half the stable length.

[0019] After clock delay extension, if the stable region on the right is short, the stable sampling point is the left side of the unstable region - half the stable length + the maximum tap length.

[0020] In the above technical solution, for applications with good temperature environment control, when only one unstable region is detected, the criterion for determining whether the unstable region is in the center is: left side of unstable region = (maximum number of taps - right side of unstable region).

[0021] In the above technical solution, for applications with good temperature environment control, when only one unstable region is detected, the criterion for determining whether the unstable region is on the left is: the left side of the unstable region < (maximum number of taps - the right side of the unstable region).

[0022] In the above technical solution, for applications with good temperature environment control, when only one unstable region is detected, the criterion for determining whether the unstable region is on the right is: left side of unstable region > (maximum number of taps - right side of unstable region).

[0023] In the above technical solution, for applications with harsh temperature environment control, the number of taps corresponding to the sampling overlapping area is m(2r-1), where m is the maximum tap position that iodelay can delay; if the length of the detected stable region is greater than m(2r-1)+3, then the stable region is usable; when a transition edge is detected, if the distance between the sampling position and the transition edge is greater than m(2r-1)+3, sampling errors can be avoided.

[0024] In the above technical solution, for applications with harsh temperature environment control, after clock delay expansion, the method for determining that the right region is longer is: the length of the first stable region detected > the length of the second stable region.

[0025] In the above technical solution, for applications with harsh temperature environment control, after clock delay extension, the method for determining that the left region is longer is: the length of the first stable region detected is less than the length of the second stable region.

[0026] In the above technical solution, the imaging system of the detector to which this fine training method is applicable includes: camera controller, imaging controller power supply chip, imaging controller, detector power supply chip, detector, erasable main and backup flash, PROM, MRAM, cameralink chip, cameralink connector, 2711 chip and 2711 connector.

[0027] The imaging controller is connected to the camera controller via the imaging controller power supply chip, to the detector via the detector power supply chip, to the cameralink connector via the cameralink chip, and to the 2711 connector via the 2711 chip; the imaging controller is also connected to the erasable main and backup flash memory, the PROM, and the MRAM respectively.

[0028] The camera controller receives a primary power input from an external source and generates the required power supply; simultaneously, it receives a second pulse input from an external source and communicates with the outside world via the 1553 bus.

[0029] The camera controller communicates with the imaging controller via a 422 communication signal and provides a second pulse to the imaging controller, and provides power to the imaging controller through the imaging controller power supply chip;

[0030] The imaging controller provides drive control signals to the detector and supplies power to the detector through the detector power supply chip, and receives serial image data output by the detector;

[0031] The imaging controller is simultaneously connected to the PROM and the erasable main and backup flash memory, which are used to load the configuration data source, and is also connected to the MRAM to update the correction coefficients and load the correction coefficients before each shooting.

[0032] The imaging controller may, depending on the selection, output image data via the CameraLink chip and the CameraLink connector, and / or via the 2711 chip and the 2711 connector.

[0033] The present invention has the following beneficial effects:

[0034] 1. When there are defects in the wiring inside the detector or the circuit board, an additional unstable region may be added on top of the two normal unstable regions. Different sampling strategies are set according to the different states of the number of unstable regions and the number of stable regions obtained by sampling, so as to obtain the optimal sampling position and improve temperature adaptability.

[0035] 2. Based on the relationship between the frequency of the serial image data clock and the reference clock frequency, dual-edge sampling is selectively used to widen the sampling area, avoid sampling overlap areas or control the sampling overlap area within the allowable range, and corresponding control is performed when setting the sampling position, thereby obtaining a wider temperature adaptation range.

[0036] 3. To address the issue that using only serial image data for delayed sampling to an unstable region makes it impossible to determine whether the nearest unstable region appears on the left or right, we propose further delaying the sampling clock on top of the serial data delay. This would provide more complete information on the distance to unstable regions, allowing for the setting of the optimal sampling position and improving temperature adaptability. Attached Figure Description

[0037] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0038] Figure 1 This is a schematic diagram of the imaging system structure of the detector to which the fine training method of the detector of the present invention is applicable.

[0039] Figure 2 This is a schematic diagram showing the state of three unstable regions and two stable regions detected by the fine training method of the detector of the present invention.

[0040] Figure 3 and Figure 4These are schematic diagrams illustrating the state of the detector, which, using the fine-training method of the detector of this invention, detected two unstable regions and one stable region. Figure 3 A schematic diagram illustrating the state of a detected long stable region; Figure 4 This is a schematic diagram illustrating the state of a short stable region being detected.

[0041] Figure 5-7 These are schematic diagrams illustrating the state where only one unstable region is detected by the present invention. Figure 5 This is a schematic diagram showing the state when the unstable region is at the exact center of the sampling region; Figure 6 This is a schematic diagram showing the state when the unstable region is to the left of the sampling region; Figure 7 This is a schematic diagram of the state when the unstable region is to the right of the sampling region.

[0042] Figure 8 This is a schematic diagram of the new sampling topology for the fine training method of the detector of the present invention. Detailed Implementation

[0043] The inventive concept of this invention is as follows: During the training process, the detector typically operates without problems when using traditional methods. However, if there are defects in the internal wiring of the detector or the circuit board, during bit correction, the serial data to be detected may have an additional unstable region on top of the normal two unstable regions. In addition to the three conventional states (no edge detection, one edge detection, and two edges detection), an extra edge position may be detected. That is, it may detect two stable length regions with three edges, or two edges with one longer or shorter stable region, or only one edge position. For applications where the frequency of the serial image data is less than half the reference clock, dual-edge sampling of the sampling clock can be used to expand the sampling area. However, for applications where the frequency of the serial image data is greater than half the reference clock, using dual-edge sampling of the sampling clock to expand the sampling area may lead to misjudgment of the sampling position due to overlapping sampling areas. Without bit width expansion, incomplete sampling states may result in poor environmental adaptability.

[0044] To address this, the present invention proposes a fine training method for a detector, which classifies the detector based on the number of stable and unstable regions sampled, and expands the sampling region by delaying the serial image data and the sampling clock respectively, thereby obtaining complete sampling information and setting the optimal sampling position.

[0045] In a specific embodiment of the present invention:

[0046] Camera Link refers to a data transmission protocol based on the Channel Link protocol, used in the field of machine vision for communication between industrial cameras and image acquisition cards.

[0047] Flash refers to non-volatile memory.

[0048] PROM stands for Programmable Read-Only Memory.

[0049] MRAM stands for Magnetic Random Access Memory.

[0050] iodelay refers to the input / output delay unit.

[0051] Maximum tap count refers to maximum latency.

[0052] The present invention will now be described in detail with reference to the accompanying drawings.

[0053] like Figure 1 As shown, the fine training method for the detector of the present invention is applicable to the imaging system of the detector, which mainly includes: a camera controller, an imaging controller power supply chip, an imaging controller, a detector power supply chip, a detector, an erasable and rewritable main and backup flash, a PROM (programmable read-only memory), an MRAM (magnetic random access memory), a cameralink chip, a cameralink connector, a 2711 chip, and a 2711 connector.

[0054] The camera controller receives a primary power supply from an external source and generates the necessary power supplies. It also receives a second pulse from an external source and communicates with external devices 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 its power supply chip. The imaging controller provides drive control signals to the multispectral detector and supplies power to the multispectral detector via its detector power supply chip, receiving serial image data output from the detector. The imaging controller is connected to both a configuration data source PROM and erasable primary / backup flash memory, and is also connected to MRAM for updating correction coefficients and loading correction coefficients before each image capture. The imaging controller can output image data either via a CameraLink chip and CameraLink connector, or via a 2711 chip and 2711 connector, depending on the desired method.

[0055] The fine training method for the detector of the present invention is described in detail below:

[0056] I. Applications requiring good temperature environment control:

[0057] 1. For example Figure 2 As shown, when three unstable regions and two stable regions are detected, the center of the region with the longer stable region is selected as the final sampling location. The number of stable taps in the longer stable region is greater than the number of taps in the shorter stable region.

[0058] 2. For example Figure 3 As shown, when a long stable region and two unstable regions are detected, and the sum of the widths of the long stable region and the unstable region is greater than half of the total number of delay taps, the sampling position is the midpoint of the long stable region, which is: the left edge of the first stable region + half the width of the long stable region.

[0059] like Figure 4 As shown, when a short stable region and two unstable regions are detected, and the sum of the widths of the long stable region and one unstable region is less than or equal to half of the total number of delay taps, the sampling location is:

[0060] Maximum number of taps + left edge of the first unstable region - (maximum number of taps - width of two unstable regions - width of the short stable region) / 2 =

[0061] The left edge of the first unstable region + half the maximum number of taps - (width of the two unstable regions + width of the short stable region) / 2.

[0062] 3. Only one unstable region was detected;

[0063] like Figure 5 As shown, when the unstable region is in the exact center, the sampling position is set to half of the left edge of the unstable region. The criterion for this is: left side of unstable region = (maximum number of taps - right side of unstable region).

[0064] like Figure 6 As shown, when the unstable region is on the left, the sampling position is set to the right edge of the unstable region + (maximum number of taps - right edge of the unstable region) / 2 = half of the maximum number of taps + half of the right edge of the unstable region. The judgment criterion is: left side of the unstable region < (maximum number of taps - right side of the unstable region).

[0065] like Figure 7 As shown, when the unstable region is on the right, the sampling position is set to half of the left edge of the unstable region. The criterion is: left side of unstable region > (maximum number of taps - right side of unstable region).

[0066] II. Applications requiring harsh temperature control environments:

[0067] 1. Obtaining a wider sampling area based on dual-edge sampling of the sampling clock.

[0068] (a) When the serial image data frequency is less than or equal to half of the reference clock frequency, the sampling area is widened by sampling based on the dual edge of the sampling clock. All sampling points do not overlap, and the maximum performance expansion can be obtained.

[0069] (b) When the serial image data frequency is greater than 0.5 times the reference clock frequency but less than r times the reference clock frequency (0.5 < r < 0.6), the sampling area is widened by sampling based on the double edge of the sampling clock. Some sampling points will overlap, which may lead to the final sampling error or the sampling is not in the optimal position. The method to reduce the influence of the overlapping area is to limit the length range of the detected stable area, and at the same time, limit the distance between the sampling position and the transition edge when a transition edge is detected.

[0070] The number of taps corresponding to the sampling overlap area is m(2r-1), where m is the maximum tap position that iodelay can delay; the length of the detected stable region is greater than m(2r-1)+3, and the stable region can be considered usable; when a transition edge is detected, the distance between the sampling position and the transition edge is greater than m(2r-1)+3 to avoid the risk of sampling error.

[0071] 2. First, delay the serial image data to find the sampling position, and then delay the sampling clock position to find the sampling position.

[0072] The novel sampling topology of this invention is as follows: Figure 8 As shown, serial image data is delayed by a controllable data delay such as iodelay, and the sampling clock is delayed by a controllable clock delay such as DCM. Finally, the data is sampled by a sampler such as a D flip-flop. Using this method, at least two stable regions and three unstable regions can typically be detected.

[0073] The time length corresponding to a single position of a controllable clock delay at p positions:

[0074]

[0075] In the formula, t data q represents the period length of the serial image data, and q represents the number of parts into which the period length of the serial image data is divided.

[0076] The time length corresponding to the controllable clock delay:

[0077]

[0078] The correspondence between the controllable clock delay position p and the controllable data delay tap number:

[0079]

[0080] In the formula, t reference is the reference clock cycle length of iodelay, and m is the maximum tap position that iodelay can delay;

[0081] (a) After clock delay extension, if the stable region on the right is long, the stable sampling point is the right side of the unstable region plus half the stable length.

[0082] The method for determining if the right-side region is longer is: the length of the first stable region detected is greater than the length of the second stable region.

[0083] (b) After clock delay extension, if the stable region on the right is short, the stable sampling point is the left side of the unstable region - half the stable length + the maximum tap length.

[0084] The method for determining if the left-side region is longer is: the length of the first stable region detected is less than the length of the second stable region.

[0085] In a specific embodiment of the present invention:

[0086] (1) The imaging controller power supply chip uses a 510 DC-DC module;

[0087] (2) The detector used is from Changguang Chenxin Company;

[0088] (3) The camera controller mainly uses DSP chips;

[0089] (4) The imaging controller mainly adopts the imaging controller and refresh chip of Shanghai Fudan Microelectronics Co., Ltd.;

[0090] (5) The detector power supply chip mainly uses TI's LDO;

[0091] (6) The 2711 chip uses the TLK2711 chip; the 2711 connector uses the micro coaxial connector from Sichuan Huafeng; the cameralink chip uses the DS90CR287; the cameralink connector uses the MDR26 connector from 3M; the erasable main and backup flash uses products from Fudan Microelectronics; the PROM uses products from Xilinx; and the MRAM uses products from the 771 Institute.

[0092] When there are defects in the wiring inside the detector or the circuit board, this invention may add an unstable region in addition to the two conventional unstable regions. Different sampling strategies are set according to the different states of the number of unstable regions and the number of stable regions obtained by sampling, so as to obtain the optimal sampling position and improve temperature adaptability.

[0093] Based on the relationship between the frequency of the serial image data clock and the reference clock frequency, this invention selectively uses dual-edge sampling to widen the sampling area, avoids sampling overlap areas or controls the sampling overlap areas within an allowable range, and performs corresponding control when setting the sampling position, thereby obtaining a wider temperature adaptation range.

[0094] This invention addresses the problem that using only serial image data for delayed sampling to an unstable region makes it impossible to determine whether the nearest unstable region appears on the left or right. It proposes to further delay the sampling clock on top of the serial data delay to obtain more complete information on the distance to unstable regions, thereby enabling the setting of the optimal sampling position and improving temperature adaptability.

[0095] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for fine-tuning a detector, characterized in that, For applications with well-controlled temperature environments, the following steps are included: When three unstable regions and two stable regions are detected, the center of the region with the longer stable region is selected as the final sampling location. When a long stable region and two unstable regions are detected, and the sum of the widths of the long stable region and the unstable region is greater than half of the total number of delay taps, the sampling position is the midpoint of the long stable region. When one short stable region and two unstable regions are detected, characterized by the sum of the widths of the short stable region and the unstable region being less than or equal to half of the total number of delay taps, the sampling location is: Maximum number of taps + left edge of the first unstable region - (maximum number of taps - width of the two unstable regions - width of the short stable region) / 2 = The left edge of the first unstable region + half of the maximum number of taps - (width of the two unstable regions + width of the short stable region) / 2; When only one unstable region is detected: When the unstable region is at the exact center, the sampling position is set to half of the left edge of the unstable region; When the unstable region is on the left, the sampling position is set to the right edge of the unstable region + (maximum number of taps - right edge of the unstable region) / 2 = half of the maximum number of taps + half of the right edge of the unstable region; When the unstable region is on the right, the sampling position is set to half of the left edge of the unstable region; For applications with harsh temperature control conditions, the following steps are included: When the serial image data frequency is less than or equal to half of the reference clock frequency, the sampling area is widened by dual-edge sampling based on the sampling clock. When the serial image data frequency is greater than 0.5 times the reference clock frequency but less than r times the reference clock frequency, 0.5 < r < 0.6, the length range of the detected stable region is limited. At the same time, when a transition edge is detected, the distance between the sampling position and the transition edge is limited. The widened sampling region is obtained based on the dual-edge sampling of the sampling clock. After clock delay extension, if the stable region on the right is long, the stable sampling point is the right side of the unstable region plus half the stable length. After clock delay extension, if the stable region on the right is short, the stable sampling point is the left side of the unstable region - half the stable length + the maximum tap length.

2. The fine training method for the detector according to claim 1, characterized in that, For applications with well-controlled temperature environments, when only one unstable region is detected, the criterion for determining whether the unstable region is in the exact center is: left side of unstable region = (maximum number of taps - right side of unstable region).

3. The fine training method for the detector according to claim 1, characterized in that, For applications with well-controlled temperature environments, when only one unstable region is detected, the criterion for determining whether the unstable region is on the left is: left side of the unstable region < (maximum number of taps - right side of the unstable region).

4. The fine training method for the detector according to claim 1, characterized in that, For applications with well-controlled temperature environments, when only one unstable region is detected, the criterion for determining whether the unstable region is on the right is: left side of the unstable region > (maximum number of taps - right side of the unstable region).

5. The fine training method for the detector according to claim 1, characterized in that, For applications with harsh temperature control, the number of taps corresponding to the sampling overlap area is m(2r-1), where m is the maximum tap position that iodelay can delay; if the length of the detected stable region is greater than m(2r-1)+3, then the stable region is usable; if a transition edge is detected, the distance between the sampling position and the transition edge is greater than m(2r-1)+3 to avoid sampling errors.

6. The fine training method for the detector according to claim 1, characterized in that, For applications with harsh temperature control conditions, after clock delay extension, the method to determine if the right-side region is longer is: the length of the first stable region detected is greater than the length of the second stable region.

7. The fine training method for the detector according to claim 1, characterized in that, For applications with harsh temperature control conditions, after clock delay extension, the method for determining if the left region is longer is: the length of the first stable region detected is less than the length of the second stable region.

8. The fine training method for the detector according to any one of claims 1-7, characterized in that, The imaging system of the detector to which this fine training method is applicable includes: camera controller, imaging controller power supply chip, imaging controller, detector power supply chip, detector, erasable main and backup flash, PROM, MRAM, cameralink chip, cameralink connector, 2711 chip and 2711 connector. The imaging controller is connected to the camera controller via the imaging controller power supply chip, to the detector via the detector power supply chip, to the cameralink connector via the cameralink chip, and to the 2711 connector via the 2711 chip; the imaging controller is also connected to the erasable main and backup flash memory, the PROM, and the MRAM respectively. The camera controller receives a primary power input from an external source and generates the required power supply; simultaneously, it receives a second pulse input from an external source and communicates with the outside world via the 1553 bus. The camera controller communicates with the imaging controller via a 422 communication signal and provides a second pulse to the imaging controller, and provides power to the imaging controller through the imaging controller power supply chip; The imaging controller provides drive control signals to the detector and supplies power to the detector through the detector power supply chip, and receives serial image data output by the detector; The imaging controller is simultaneously connected to the PROM and the rewritable primary and backup flash memory to load the configuration data source, and is also connected to the MRAM to update the correction coefficients and load the correction coefficients before each shooting. The imaging controller may, depending on the selection, output image data via the CameraLink chip and the CameraLink connector, and / or via the 2711 chip and the 2711 connector.

Images