A real-time processing method and device of a collimator

Abstract

The application provides a real-time processing method and device of a self-collimating light pipe, and solves the technical problem of the efficiency defect in the signal processing process of the existing self-collimating light pipe. The method comprises the following steps: determining a CMOS image sensor windowing range to form an effective pixel acquisition matrix according to a preset slit image parameter; performing sensor line-by-line exposure within the effective pixel acquisition matrix range, and synchronously transmitting image data collected by a previous line of sensors during the exposure process; setting at least two memory buffer areas to cyclically receive each frame of imaging data formed by the effective pixel acquisition matrix; and performing image dimension reduction on each frame of imaging to form corresponding one-dimensional image data. The frame frequency of the sensor is improved, the waiting time between the processing processes is reduced, the utilization efficiency of the processor is improved, and the real-time performance and reliability of the self-collimating light pipe signal output are ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a real-time processing method and device of a collimator. BACKGROUND

[0002] The existing technology's aiming system is mainly composed of an inertial north-seeking device and a collimator. The collimator can calculate the azimuth deviation angle of the collimator and the target and provide high-precision aiming directional information. The system composition of the collimator is shown in Figure 1 . In the Figure 1 , the collimator mainly includes an optical system combination for light signal transmission and reception, a CMOS image sensor for receiving analog-to-digital conversion and front-end signal processing of the light signal, a main control board for data processing and forwarding control of the collected image, and a bidirectional communication link formed by the main control board and an upper computer or a laser strapdown inertial device, such as a network port or an RS-422 serial port, to realize data transmission.

[0003] In view of the high real-time requirement of the aiming system, improving the calculation frequency of the collimator and the output precision of the collimation deviation angle is the basis for improving the use reliability of the aiming system. The core device inside the collimator is a CMOS image sensor. Improving the frame frequency of the CMOS image sensor and reducing the operation time of a single image can improve the real-time performance and reliability of the output data of the collimator, which is a feasible technical idea. SUMMARY

[0004] In view of the above problems and technical ideas, the present application provides a real-time processing method and device of a collimator, which solves the technical problem of the efficiency defect in the signal processing process of the existing collimator.

[0005] The real-time processing method of the collimator according to the present application comprises the following steps.

[0006] Determine the windowing range of the CMOS image sensor according to the preset slit image parameters to form an effective pixel acquisition matrix;

[0007] Perform row-by-row exposure of the sensor within the range of the effective pixel acquisition matrix, and synchronously transmit the image data collected by the previous row of sensors during the exposure process;

[0008] Set at least two memory buffer areas to cyclically receive each frame of imaging data formed by the effective pixel acquisition matrix;

[0009] Perform image dimension reduction on each frame of imaging to form corresponding one-dimensional image data.

[0010] In an embodiment of the present application, the formation of the effective pixel acquisition matrix comprises the following steps.

[0011] The number of effective pixel rows of the collimating slit is determined according to the size of the collimating slit and the photosensitive unit of the CMOS image sensor;

[0012] The effective pixel acquisition matrix in the windowing range of the CMOS image sensor is determined in combination with the number of effective pixel rows of the collimating slit image;

[0013] The acquisition frame frequency is determined according to the working frequency of the main control board and the effective pixel acquisition matrix.

[0014] In an embodiment of the present application, the image data collected by the previous row of sensors during the exposure process comprises:

[0015] The photosensitive units are configured by the row register within the range of the effective pixel acquisition matrix, and the photosensitive units are exposed row by row;

[0016] When the value of the row register is stepped, the image information formed by the exposure of the previous row of photosensitive units is transmitted.

[0017] In an embodiment of the present application, the cyclically received imaging data of each frame formed by the effective pixel acquisition matrix comprises:

[0018] Three memory buffer areas are set for storing three frames of complete imaging data, and the latest frame of imaging data is sequentially stored into the three memory buffer areas;

[0019] The address number of the memory buffer area that is newly completed is obtained through the access channel, and the memory copy function is used to copy the to-be-processed image data from the memory buffer area corresponding to the address number to the image processing memory according to the preset parameters.

[0020] In an embodiment of the present application, the image dimensionality reduction comprises:

[0021] The arithmetic mean value of the pixels in the column is obtained by traversing the to-be-processed image data column by column, and the one-dimensional image data after fitting of a row is obtained by using the arithmetic mean value.

[0022] The real-time processing device of the self-collimating light pipe in the embodiment of the present application comprises:

[0023] The frame frequency improvement module is configured to determine the effective pixel acquisition matrix formed by the windowing range of the CMOS image sensor according to preset slit image parameters;

[0024] The data transmission module is configured to perform row-by-row exposure of the sensor within the range of the effective pixel acquisition matrix, and synchronously transmit the image data collected by the previous row of sensors during the exposure process.

[0025] The image buffer module is configured to set at least two memory buffer areas for cyclically receiving imaging data of each frame formed by the effective pixel acquisition matrix.

[0026] The dimension reduction processing module is configured to perform image dimension reduction on each frame of imaging to form corresponding one-dimensional image data.

[0027] In an embodiment of the present application, the frame rate improvement module comprises:

[0028] The key acquisition unit is configured to determine the number of valid pixel rows of the collimating slit image according to the size of the collimating slit and the photosensitive unit of the CMOS image sensor.

[0029] The acquisition windowing unit is configured to determine the valid pixel acquisition matrix in the windowing range of the CMOS image sensor in combination with the number of valid pixel rows of the collimating slit image.

[0030] The frame rate improvement unit is configured to determine the acquisition frame rate according to the working frequency of the main control board and the valid pixel acquisition matrix.

[0031] In an embodiment of the present application, the data transmission module comprises:

[0032] The row-by-row exposure unit is configured to perform row-by-row exposure of the photosensitive unit in the range of the valid pixel acquisition matrix through the row register configuration.

[0033] The row-by-row transmission unit is configured to transmit the image information formed by exposure of the previous row of photosensitive unit when the value of the row register is stepped.

[0034] In an embodiment of the present application, the image buffer module comprises:

[0035] The sequential buffering unit is configured to set three memory buffer areas for storing three frames of complete imaging data, and the latest frame of imaging data is sequentially and alternately stored in the three memory buffer areas.

[0036] The memory reading unit is configured to obtain the address number of the memory buffer area that is newly completed and buffered through an access channel, and copy the to-be-processed image data from the memory buffer area corresponding to the address number to the image processing memory according to a preset parameter using a memory copy function.

[0037] In an embodiment of the present application, the dimension reduction processing module comprises:

[0038] The dimension reduction processing unit is configured to obtain the arithmetic mean value of the pixels in the column by traversing the to-be-processed image data column by column, and obtain one-dimensional image data after fitting of one row by using the arithmetic mean value.

[0039] The real-time processing method and device of the self-collimating light pipe according to the embodiment of the present application reduce the single-frame imaging output data of the CMOS image sensor by improving the effective pixel ratio, improve the frame frequency of the sensor, realize real-time acquisition of the latest image data during image processing through the ordered caching of each frame of imaging data, reduce the waiting time between the processing processes, reduce the computing power demand during image processing by reducing the dimensionality of the imaging data after the windowing processing, improve the utilization efficiency of the processor, and ensure the real-time performance and reliability of the signal output of the self-collimating light pipe. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 Fig. 1 shows the schematic diagram of the architecture of the self-collimating light pipe in the prior art.

[0041] Figure 2 Fig. 2 shows the flowchart of the real-time processing method of the self-collimating light pipe according to an embodiment of the present application.

[0042] Figure 3 Fig. 3 shows the application diagram of the real-time processing method of the self-collimating light pipe according to an embodiment of the present application.

[0043] Figure 4 Fig. 4 shows the schematic diagram of the architecture of the real-time processing device of the self-collimating light pipe according to an embodiment of the present application. DETAILED DESCRIPTION

[0044] To make the objectives, technical solutions, and advantages of the present application clearer and more apparent, the present application is further described below in combination with the drawings and specific embodiments. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0045] The real-time processing method of the self-collimating light pipe according to an embodiment of the present application is shown in Fig. 2. Figure 2 In the embodiment, the present embodiment includes: Figure 2

[0046] Step 100: Determine the windowing range of the CMOS image sensor according to the preset slit image parameters to form an effective pixel acquisition matrix.

[0047] Those skilled in the art can understand that the slit image refers to the image formed through the optical system. By adjusting the optical system, the effective light-sensing imaging range can be limited on the two-dimensional light-sensing unit matrix of the CMOS image sensor, and the windowing mode is formed in cooperation with the register configuration of the CMOS image sensor, thereby improving the effective pixel ratio of the image returned by the optical system and reducing the processing overhead of the invalid pixels.

[0048] ​Step 200: performing sensor line-by-line exposure within the range of the effective pixel acquisition matrix, and synchronously transmitting the image data acquired by the previous line of sensors during the exposure process.

[0049] The exposure process of the photosensitive unit is formed by performing sensor line-by-line exposure, and when the current line of sensors is exposed, the synchronous transmission of the image information formed by the exposure of the previous line of photosensitive units is performed.

[0050] Step 300: setting at least two memory buffer areas for cyclically receiving each frame of imaging data formed by the effective pixel acquisition matrix.

[0051] By setting a plurality of cyclically used memory buffer areas to receive each frame of exposure image formed by the effective pixel acquisition matrix, the exposure image buffer is provided for subsequent image processing, the continuity of the exposure image acquisition and the continuity of the exposure image processing are met, and the waiting time of the handshake signal between the two processes is eliminated.

[0052] Step 400: performing image dimension reduction on each frame of imaging to form corresponding one-dimensional image data.

[0053] The exposure image processing process reduces the algorithm consumption of data processing by reducing the dimension of each frame of two-dimensional image data to form one-dimensional image data. The image dimension reduction process can extract the main feature components in the exposure image by using a linear transformation process.

[0054] The real-time processing method of the self-collimating light pipe according to the embodiment of the application reduces the single-frame imaging output data of the CMOS image sensor by improving the effective pixel ratio, improves the frame frequency of the sensor, realizes real-time acquisition of the latest image data during image processing by orderly buffering each frame of imaging data, reduces the waiting time between the processing processes, reduces the algorithm requirement by reducing the dimension of the imaging data after the windowing process during image processing, improves the utilization efficiency of the processor, and ensures the real-time performance of the self-collimating light pipe signal.

[0055] The application of the real-time processing method of the self-collimating light pipe according to an embodiment of the application is shown in Figure 3 The application of the real-time processing method of the self-collimating light pipe according to an embodiment of the application is shown in Figure 2 In the embodiment of the application, step 100 includes:

[0056] Step 110: determining the number of effective pixel rows of the collimating slit according to the size of the collimating slit and the photosensitive unit of the CMOS image sensor.

[0057] In the embodiment of the application, the size of the CMOS image sensor is 7.5 um, the collimating slit occupies 170 rows, and the number of effective pixel rows of the collimating slit is determined as 576 on the basis of considering leaving a proper margin for the installation of the optical system, and the balance of the number of rows of a single picture is about 3 times the height of the slit, and the minimum stepping distance of the sensor and other sensor parameters.

[0058] Step 120: Determine the effective pixel acquisition matrix within the window range of the CMOS image sensor in combination with the number of effective pixel rows of the collimated slit image.

[0059] The number of effective pixel rows in the collimated slit image determines the range of photosensitive cells in the pixel data set for collimation analysis, forming a windowing range within the CMOS image sensor's photosensitive cell matrix. This, in turn, determines the effective pixel acquisition matrix corresponding to the windowing range. The effective pixel acquisition matrix range can be controlled using the CMOS image sensor's windowing register. Depending on the CMOS image sensor model, the windowing register can be a single-function register or a combination of other basic register circuit functions.

[0060] In one embodiment of the present invention, the above windowing mode can increase the effective pixel ratio in a single picture from 4.88% to 26.04%.

[0061] Step 130: Determine the acquisition frame rate according to the main control board operating frequency and the effective pixel acquisition matrix.

[0062] In one embodiment of the present invention, the operating frequency of the main control board is 25 MHz, and the line period T L =T m *n,T m is the main frequency period, n is the comprehensive configuration coefficient of TX signal, sampling signal and PGA related registers, which is 263, that is, T L =10.52us, the number of output lines is 576, and the single-frame output time is T p =T L *Line=6.05952ms, frame rate Meet system requirements.

[0063] The real-time processing method for an autocollimating light tube according to the present invention uses a windowing pattern to determine the effective pixel acquisition matrix of the photosensitive unit, forming an imaging frame output that is closely correlated with the collimation signal analysis. This reduces the frame output data volume, significantly increasing the output frame rate within a limited output bandwidth.

[0064] like Figure 2 As shown, in one embodiment of the present invention, step 200 includes:

[0065] Step 210: configuring the photosensitive units through the row registers within the effective pixel acquisition matrix range, and performing row-by-row exposure of the photosensitive units.

[0066] Step 220: When the row register value is stepped, the image information formed by the exposure of the previous row of photosensitive units is transmitted.

[0067] The self-collimating light pipe real-time processing method of the embodiment of the present application utilizes a row-by-row exposure process to form a synchronous transmission process, avoids data processing delay caused by data transmission, and effectively optimizes the transmission efficiency of exposure data.

[0068] As shown in the figure, Figure 2 In an embodiment of the present application, step 300 includes:

[0069] Step 310: Three memory cache areas are set for storing three frames of complete imaging data, and the latest frame of imaging data is sequentially stored in the three memory cache areas.

[0070] The CMOS image sensor raw data storage addresses are 0x10000000, 0x11000000 and 0x12000000, and the latest image is sequentially stored in the three addresses, forming a data read-write ping-pong mechanism.

[0071] Step 320: The address number of the latest completed cache memory cache area is obtained through an access channel, and the mth to nth row of image data to be processed is copied to the image processing memory according to a preset parameter using a memory copy function.

[0072] In an embodiment of the present application, the address number query function of the memory cache area for storing the latest image is executed through an interrupt mechanism. According to a preset parameter, the mth to nth row of data in the memory cache area is copied to the memory (which can be part or all of the complete exposure image) using the memcpy function.

[0073] The self-collimating light pipe real-time processing method of the embodiment of the present application can adapt to the timely caching of high-frame-rate exposure images, provide a fast image reading mechanism for the image processing process, and provide an on-demand reading process for complete image data for the image processing process. The cache reading efficiency of exposure data is effectively improved. Figure 2 As shown in the figure,

[0074] Step 410: The arithmetic mean value of the pixels in the column is obtained by traversing the image data to be processed column by column, and a one-dimensional image data after fitting is obtained using the arithmetic mean value.

[0075] The self-collimating light pipe real-time processing method of the embodiment of the present application uses a dimension reduction algorithm to simplify the data processing process, so that the subsequent algorithms of the host board can fully utilize the limited computing power on the basis of dimension reduction data, and the reliability of data processing is guaranteed.

[0076] An embodiment of the self-collimating light pipe real-time processing device of the present application comprises:

[0077] The memory is used to store the program code in the processing process of the self-collimating light pipe real-time processing method of the above-mentioned embodiment.

[0078] A processor is configured to execute program codes in the process of the real-time processing method of the self-collimation light tube.

[0079] The processor can be a Digital Signal Processor (DSP), a Field-Programmable Gate Array (FPGA), a Microcontroller Unit (MCU), a system on a chip (SoC), or a Programmable Logic Controller (PLC) minimum system including I / O.

[0080] An embodiment of the real-time processing device of the self-collimation light tube is shown in Figure 4 In Figure 4 , the embodiment includes:

[0081] A frame rate improvement module 10 is configured to determine an effective pixel acquisition matrix in a CMOS image sensor window range according to preset slit image parameters.

[0082] A data transmission module 20 is configured to perform sensor line-by-line exposure in the effective pixel acquisition matrix range and synchronously transmit image data collected by a previous line of sensors during the exposure process.

[0083] An image buffer module 30 is configured to set at least two memory buffer areas to cyclically receive each frame of imaging data formed by the effective pixel acquisition matrix.

[0084] A dimension reduction processing module 40 is configured to perform image dimension reduction on each frame of imaging to form corresponding one-dimensional image data.

[0085] As shown in Figure 4 , in an embodiment of the present application, the frame rate improvement module 10 includes:

[0086] A key acquisition unit 11 is configured to determine the number of effective pixel rows of a collimation slit image according to the size of a collimation slit and a CMOS image sensor photosensitive unit.

[0087] An acquisition window unit 12 is configured to determine an effective pixel acquisition matrix in a CMOS image sensor window range in combination with the number of effective pixel rows of a collimation slit image.

[0088] A frame rate improvement unit 13 is configured to determine an acquisition frame rate according to a main control board working frequency and the effective pixel acquisition matrix.

[0089] As shown in Figure 4 , in an embodiment of the present application, the data transmission module 20 includes:

[0090] A row-by-row exposure unit 21 is used to configure the photosensitive units within the effective pixel acquisition matrix through the row registers and perform row-by-row exposure of the photosensitive units;

[0091] The line-by-line transmission unit 22 is used to transmit the image information formed by the exposure of the photosensitive units in the previous line when the line register value is stepped.

[0092] like Figure 4 As shown, in one embodiment of the present invention, the image cache module 30 includes:

[0093] The sequential caching unit 31 is used to set three memory cache areas for storing three frames of complete imaging data, and the latest frame imaging data is stored in the three memory cache areas in turn;

[0094] The memory reading unit 32 is used to obtain the address number of the memory buffer area that has been cached the most recently through the access channel, and use the memory copy function according to preset parameters to copy m to n lines of image data to be processed from the memory buffer area corresponding to the address number to the image processing memory.

[0095] like Figure 4 As shown, in one embodiment of the present invention, the dimensionality reduction processing module 40 includes:

[0096] The dimensionality reduction processing unit 41 is used to traverse the image data to be processed column by column to obtain the arithmetic mean of the pixels in the column, and use the arithmetic mean to obtain a row of fitted one-dimensional image data.

[0097] The above description is merely 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 a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A real-time processing method for a self-collimating light pipe, characterized in that, The method comprises the following steps: determining a CMOS image sensor windowing range according to preset slit image parameters to form an effective pixel acquisition matrix; performing sensor line-by-line exposure within the effective pixel acquisition matrix range, and synchronously transmitting image data collected by a previous line of sensors during the exposure process; setting at least two memory buffer areas to cyclically receive each frame of imaging data formed by the effective pixel acquisition matrix; performing image dimension reduction on each frame of imaging to form corresponding one-dimensional image data; the step of determining the CMOS image sensor windowing range according to the preset slit image parameters to form the effective pixel acquisition matrix comprises the following steps: determining the number of effective pixel rows of a collimating slit image according to the size of the collimating slit and the photosensitive unit of the CMOS image sensor; determining the effective pixel acquisition matrix within the CMOS image sensor windowing range in combination with the number of effective pixel rows of the collimating slit image; determining the acquisition frame frequency according to the working frequency of a main control board and the effective pixel acquisition matrix; the step of synchronously transmitting the image data collected by the previous line of sensors during the exposure process comprises the following steps: performing photosensitive unit line-by-line exposure within the effective pixel acquisition matrix range by configuring the photosensitive unit through a line register; when the value of the line register is stepped, transmitting the image information formed by the exposure of the previous line of photosensitive units; the step of performing image dimension reduction comprises the following steps: obtaining the arithmetic mean value of the pixels in the column by traversing the to-be-processed image data column by column, and obtaining one-dimensional image data after fitting a row by using the arithmetic mean value.

2. The real-time processing method of a collimator according to claim 1, wherein, the step of cyclically receiving each frame of imaging data formed by the effective pixel acquisition matrix comprises the following steps: setting three memory buffer areas to store three frames of complete imaging data, and sequentially storing the latest frame of imaging data into the three memory buffer areas; obtaining the address number of the memory buffer area that has been newly completed by accessing a channel, and copying the to-be-processed image data from the memory buffer area corresponding to the address number into an image processing memory by using a memory copy function according to preset parameters.

3. A real-time processing device for a collimator, characterized in that The method comprises the following steps: a frame frequency improvement module is configured to determine a CMOS image sensor windowing range according to preset slit image parameters to form an effective pixel acquisition matrix; a data synchronous transmission module is configured to perform sensor line-by-line exposure within the effective pixel acquisition matrix range, and synchronously transmit image data collected by a previous line of sensors during the exposure process; an image buffer module is configured to set at least two memory buffer areas to cyclically receive each frame of imaging data formed by the effective pixel acquisition matrix; a dimension reduction processing module is configured to perform image dimension reduction on each frame of imaging to form corresponding one-dimensional image data; the frame frequency improvement module comprises the following units: a key acquisition unit is configured to determine the number of effective pixel rows of a collimating slit image according to the size of the collimating slit and the photosensitive unit of the CMOS image sensor; an acquisition windowing unit is configured to determine the effective pixel acquisition matrix within the CMOS image sensor windowing range in combination with the number of effective pixel rows of the collimating slit image; a frame frequency improvement unit is configured to determine the acquisition frame frequency according to the working frequency of a main control board and the effective pixel acquisition matrix; the data synchronous transmission module comprises the following units: a line-by-line exposure unit is configured to perform photosensitive unit line-by-line exposure within the effective pixel acquisition matrix range by configuring the photosensitive unit through a line register; a line-by-line transmission unit is configured to, when the value of the line register is stepped, transmit the image information formed by the exposure of the previous line of photosensitive units; the dimension reduction processing module comprises the following units: The dimension reduction processing unit is used for traversing the image data to be processed column by column to obtain the arithmetic mean value of the column pixels, and obtaining one-dimensional image data after fitting a row by using the arithmetic mean value.

4. The real-time processing method of a collimator according to claim 3, wherein, The image buffer module comprises: The sequential buffer unit is used for setting three memory buffer areas for storing three frames of complete imaging data, and sequentially storing the latest frame of imaging data into the three memory buffer areas; The memory reading unit is used for obtaining the address number of the memory buffer area which is newly completed and buffered through an access channel, and copying the image data to be processed from the memory buffer area corresponding to the address number to the image processing memory according to a preset parameter using a memory copy function.

Images