Multi-step Histogram Peak Extraction Method and System for Laser Ranging

Abstract

The present invention provides a multi-step histogram peak extraction method and system for laser ranging. By dividing the bit width N of the measurement data into several set step lengths, the peak extraction range of the measurement data is dynamically reduced. Specifically, a histogram is generated with the set step length. After storing the abscissa corresponding to the peak, the candidate range is determined by the abscissa corresponding to the peak. The measurement data that meets the candidate range is selected, and the next set step length is updated to the current set step length. The process of dynamically reducing the peak extraction range is looped until the final selection histogram is generated with the last set step length. After storing the abscissa corresponding to the final peak, the values of all stored abscissas are concatenated to obtain the final photon flight time measurement value. By dynamically reducing the peak extraction range and setting multiple variable set step lengths, the present invention can rapidly reduce the amount of data to be statistically analyzed, greatly shorten the histogram peak extraction time, and improve the peak extraction efficiency.

Description

Technical Field

[0001] The present invention relates to a multi-step histogram peak extraction method and system applied to laser ranging, belonging to the field of digital signal processing. Background Art

[0002] The main steps of histogram peak extraction include histogram formation and peak extraction. A histogram consists of an abscissa and an ordinate. Each abscissa value represents a different photon flight time, and each ordinate value corresponds to the number of occurrences of each photon flight time. The function of peak extraction is to determine a set of valid photon flight times among several groups of photon flight times, and calculate the specific distance through the valid photon flight times.

[0003] The entire laser ranging (TOF, Time of flight) system includes two steps: data acquisition and histogram peak extraction. The first step of numerical acquisition mainly comes from sensors such as time-to-digital conversion. The speed of data acquisition is generally related to the sensor structure and circuit design. There are mainly two methods for the second step of histogram peak extraction.

[0004] One of them is implemented outside the MCU (microcontroller unit). It is necessary to store several groups of photon flight time data obtained in the first step in the memory circuit in advance. During the storage process, the memory address needs to be corresponded to the photon flight time data one by one. Each time a photon flight time data is transmitted to the memory in the data acquisition part, the counter adds one to the corresponding memory address. Finally, all the count values in the memory are extracted through an algorithm, and the memory address corresponding to the histogram ordinate value with the most occurrences is found, so as to determine the photon flight time and finally calculate the measured distance. This method first counts and then extracts, and is processed in two steps, consuming a lot of time. At the same time, with the continuous upgrade of applications, the requirement for data volume is getting higher and higher, and the width and depth of the memory increase geometrically. Therefore, this method has great limitations.

[0005] Another method is to lock the histogram peak region by integrating the histogram data on the basis of data forming a histogram, and then extract the histogram peak data. However, the MCU integration operation adds the areas of each time interval (the time interval is defined as bin, which is the minimum time accuracy of the histogram). In essence, it does not significantly reduce the operation. At the same time, the limited decimal places limit the integration accuracy, which instead affects the accuracy of histogram peak extraction.

[0006] Therefore, it is very necessary to propose a method that can shorten the histogram peak extraction time on the premise of ensuring the measurement accuracy. The above problems should be considered and solved in the process of histogram peak extraction in laser ranging. Summary of the Invention

[0007] The object of the present invention is to provide a multi-step histogram peak extraction method and system for laser ranging, so as to solve the problems of long histogram peak extraction time and insufficient accuracy in the prior art.

[0008] The technical solution of the present invention is as follows:

[0009] A multi-step histogram peak extraction method for laser ranging, comprising:

[0010] Receiving the acquisition data of a time-to-digital conversion sensor and performing filtering to obtain measurement data; according to the quantity and distribution of the measurement data after noise filtering, dividing the bit width N of the measurement data into several variable set step lengths, and performing dynamic reduction of the peak extraction range on the measurement data. Specifically, generating a histogram with the set step length, obtaining the peak in the histogram, storing the abscissa corresponding to the peak, determining the candidate range from the abscissa corresponding to the peak, selecting the measurement data that meets the candidate range, and updating the next set step length to the current set step length; repeating the dynamic reduction of the peak extraction range process until the final candidate histogram is generated with the last set step length, extracting the final peak, storing the abscissa corresponding to the final peak, and splicing the values of all stored abscissas to obtain the final photon flight time measurement value.

[0011] Further, it specifically includes the following steps:

[0012] S1. Receiving the acquisition data of a time-to-digital conversion sensor and performing filtering, extracting the data in the acquisition data that is higher than the preset noise threshold to obtain measurement data, including photon flight time and photon flight times, and storing them in histogram memory one. At the same time, according to the quantity and distribution of the noise-filtered data, dividing the bit width N of the measurement data into n variable set step lengths (1…i…n) from high to low in order;

[0013] S2. Performing dynamic reduction of the peak extraction range on the measurement data, generating a candidate histogram with the set step length i and performing candidate peak extraction; statistically analyzing the measurement data in histogram memory one according to the set step length i in histogram memory two, generating a candidate histogram in histogram memory two, obtaining the first peak with the most photon flight times in the candidate histogram, storing the value of the abscissa X i at the PEAKi address of the peak memory, initializing histogram memory two, determining the candidate range where the peak is located from the value of the abscissa X i , selecting the measurement data in histogram memory one that meets the candidate range and storing it in histogram memory two, and then initializing histogram memory one; setting i = i + 1, updating the next set step length to the current set step length. If i < n, proceed to the next step S3; otherwise, proceed to step S4;

[0014] S3. Statistically analyze the measurement data in the histogram memory two obtained in step S2 in the histogram memory one according to the set step size i to generate a reselected histogram, obtain the second peak value of the most frequent photon flight times in the reselected histogram, and store the value of the abscissa X corresponding to the second peak value at the PEAKi address in the PEAK memory. Then initialize the histogram memory one. Determine the candidate range where the peak is located based on the value of the abscissa X. After selecting the measurement data that meets the candidate range in the histogram memory two and storing it in the histogram memory one, initialize the histogram memory two. Let i = i + 1, update the next set step size to the current set step size. If i < n, return to step S2; otherwise, proceed to step S4. i Store the value of the abscissa X at the PEAKi address in the PEAK memory, and then initialize the histogram memory one. Determine the candidate range where the peak is located based on the value of the abscissa X. i After selecting the measurement data that meets the candidate range in the histogram memory two and storing it in the histogram memory one, initialize the histogram memory two. Let i = i + 1, update the next set step size to the current set step size. If i < n, return to step S2; otherwise, proceed to step S4.

[0015] S4. Generate a final selection histogram with the set step size i and perform final selection peak extraction. Statistically analyze the measurement data in the histogram memory two obtained in step S2 in the histogram memory one according to the set step size i, or statistically analyze the measurement data in the histogram memory one obtained in step S3 in the histogram memory two according to the set step size i to generate a final selection histogram. Obtain the final selection peak value of the most frequent photon flight times in the final selection histogram, store the value of the abscissa Xi corresponding to the final selection peak value at the PEAKi address in the PEAK memory, and splice the values of all the abscissas stored in the PEAK memory to obtain the final photon flight time measurement value.

[0016] Further, in step S2, generating a candidate histogram with the set step size i specifically means

[0017] In step S2, generating a candidate histogram with the set step size i specifically means

[0018] S21. Set the length of the step size i to k, determine the division range as 2 N / 2 k , and divide the photon flight times of the collected measurement data into 2 k groups according to the division range.

[0019] S22. Calculate the sum of the photon flight times in each group obtained in step S21.

[0020] S23. Establish a candidate histogram for each group obtained in step S21. The abscissa of the candidate histogram is the serial number of the group, and the ordinate is the photon flight times.

[0021] Further, in step S2, determining the candidate range where the peak is located based on the value of the abscissa X specifically means i using the group where the value of the abscissa X corresponding to the first peak value is located as the candidate range. The number of measurement data within the candidate range is 2 i / 2 N / 2 k, update N = N - k.

[0022] Further, in step S4, the numerical values of all abscissas stored in the PEAK memory are spliced to obtain the final photon flight time measurement value. Specifically, after converting the lengths of the set step sizes corresponding to the abscissa numerical values at the PEAK1 address, PEAK2 address... PEAKi address into binary data, and arranging and splicing them in the order of the PEAK1 address, PEAK2 address... PEAKi address, the binary numerical value of the final photon flight time measurement value can be obtained.

[0023] Further, in step S1, the variable set step sizes are either unequal or equal.

[0024] A three-step histogram peak extraction system for laser ranging that implements the three-step histogram peak extraction method for laser ranging described in any one of the above, including a time-to-digital conversion sensor and a microcontroller unit MCU. The microcontroller unit MCU includes a control module, a noise filtering module, a peak extraction module, a histogram memory one, a histogram memory two, and a peak memory.

[0025] The microcontroller unit MCU enables the time-to-digital conversion sensor, and the time-to-digital conversion sensor starts to collect data and sends the collected data to the microcontroller unit MCU. In the microcontroller unit MCU, after the noise is filtered by the noise filtering module to obtain measurement data, the measurement data is sent to the control module. The control module divides the bit width N of the measurement data into several variable set step sizes, performs a dynamic reduction of the peak extraction range processing on the measurement data, and alternately performs peak extraction multiple times through the peak extraction module, the histogram memory one, and the histogram memory two to achieve histogram peak extraction for each set step size after the dynamic reduction of the peak extraction range. The abscissas corresponding to the extracted peaks are stored in the peak memory, and the numerical values of all abscissas stored in the peak memory are spliced to obtain the final photon flight time measurement value.

[0026] The beneficial effects of the present invention are:

[0027] First, compared with the existing two-step method of histogram generation and peak extraction for laser ranging, on the one hand, the method of the present invention can make the values to be statistically reduced at the fastest speed through the dynamic reduction of the peak extraction range and the setting of variable step sizes. On the other hand, the present invention performs histogram peak extraction while generating the histogram, greatly shortening the histogram peak extraction time and effectively improving the peak extraction efficiency.

[0028] Second, the multi-step histogram peak extraction method and system applied to laser ranging. Compared with the integral histogram extraction method, in the data processing process of the present invention, all data are utilized, and the data irrelevant to the histogram peak are selectively removed to ensure the accuracy of histogram peak extraction.

[0029] Third, the method of the present invention. Compared with the method that needs to read and write data to an external large memory, in terms of structure, the present invention continuously reuses the storage hardware resources of the main control, simplifies the system composition, and improves the utilization rate of hardware resources.

[0030] Fourth, the multi-step histogram peak extraction method and system applied to laser ranging. Compared with the existing histogram extraction methods, in the first step of the present invention, pre-filtering is performed to reduce the influence of noise on the histogram peak extraction speed. At the same time, the noise threshold will change according to the change of the measurement environment, improving the anti-interference ability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is a schematic flow chart of the multi-step histogram peak extraction method applied to laser ranging in an embodiment of the present invention.

[0032] Figure 2 is a schematic diagram for explaining the acquisition data of the time-to-digital converter in the embodiment.

[0033] Figure 3 is a schematic diagram for explaining a specific example in which the bit width of the measurement data is divided into three set step lengths in the embodiment;

[0034] Figure 4 is a schematic diagram for explaining the generation of candidate histograms with three set step lengths for peak extraction in the embodiment. Among them, (a) is a schematic diagram for explaining the generation of a candidate histogram with set step length 1, (b) is a schematic diagram for explaining the generation of a candidate histogram with set step length 2, and (c) is a schematic diagram for explaining the generation of a candidate histogram with set step length 3.

[0035] Figure 5 is a schematic diagram for explaining the experimental simulation results of the multi-step histogram peak extraction method applied to laser ranging in the embodiment.

[0036] Figure 6 is a schematic diagram for explaining the multi-step histogram peak extraction system applied to laser ranging in the embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0037] The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0038] Embodiment

[0039] A multi-step histogram peak extraction method applied to laser ranging, asFigure 1 , including:

[0040] Receiving the acquisition data of the time-to-digital conversion sensor and performing filtering to obtain measurement data; dividing the bit width N of the measurement data into several variable set steps according to the quantity and distribution of the measurement data after noise filtering, and performing dynamic peak extraction range reduction processing on the measurement data. Specifically, a histogram is generated with the set step, the peak in the histogram is obtained, the abscissa corresponding to the peak is stored, the candidate range is determined by the abscissa corresponding to the peak, the measurement data meeting the candidate range is selected, and the next set step is updated to the current set step; the dynamic peak extraction range reduction processing is performed in a loop until the last set step generates the final selection histogram, the final peak is extracted, the abscissa corresponding to the final peak is stored, and the values of all stored abscissas are concatenated to obtain the final photon flight time measurement value.

[0041] This multi-step histogram peak extraction method applied to laser ranging, compared with the existing two-step method of histogram generation and peak extraction, on the one hand, the method of the present invention can rapidly reduce the amount of data to be statistically analyzed through the dynamic reduction of the peak extraction range and the setting of variable steps; on the other hand, the present invention performs histogram peak extraction while generating the histogram, greatly shortening the histogram peak extraction time and effectively improving the peak extraction efficiency.

[0042] This multi-step histogram peak extraction method applied to laser ranging specifically includes the following steps

[0043] S1. Receiving the acquisition data of the time-to-digital conversion sensor and performing filtering, extracting the acquisition data higher than the preset noise threshold to obtain measurement data, including photon flight time and photon flight times, and storing them in the first histogram memory. The bit width N of the measurement data is sequentially divided into n variable set steps (1…i…n) from high to low; in step S1, the variable set steps are such that each set step is either unequal or equal. The noise threshold can be adjusted and set in real time according to the working environment of the actual sensor for detecting photon flight time.

[0044] S2. Performing dynamic peak extraction range reduction processing on the measurement data, generating a candidate histogram with the set step i, and performing candidate peak extraction; statistically analyzing the measurement data in the first histogram memory according to the set step i in the second histogram memory, generating a candidate histogram in the second histogram memory, obtaining the first peak with the most photon flight times in the candidate histogram, and storing the value of the abscissa X i in the PEAKi address of the peak memory, and then initializing the second histogram memory. From the abscissa X iDetermine the candidate range where the peak is located according to the value of , after selecting the measurement data that meets the candidate range in the histogram memory one and storing it in the histogram memory two, initialize the histogram memory one; let i = i + 1, update the next set step size to the current set step size, if i < n, go to the next step S3; otherwise, go to step S4;

[0045] In step S2, generate a candidate histogram with the set step size i. Specifically,

[0046] S21. Set the length of the set step size i to k, and determine the division range as 2 N / 2 k Divide the photon flight time of the collected measurement data into 2 k groups according to the division range;

[0047] S22. Calculate the sum of the photon flight times in each group obtained in step S21;

[0048] S23. Establish a candidate histogram for each group obtained in step S21. The abscissa of the candidate histogram is the serial number of the group, and the ordinate is the photon flight time.

[0049] In step S2, determine the candidate range where the peak is located according to the value of the abscissa X i Specifically, take the group where the value of the abscissa X i corresponding to the first peak is located as the candidate range. The number of measurement data within the candidate range is 2 N / 2 k Update N = N - k.

[0050] S3. Statistically process the measurement data in the histogram memory two obtained in step S2 according to the set step size i in the histogram memory one to generate a reselected histogram, obtain the second peak with the most photon flight times in the reselected histogram, store the value of the abscissa X i corresponding to the second peak at the PEAKi address in the PEAK memory, and then initialize the histogram memory one; determine the candidate range where the peak is located according to the value of the abscissa X i Select the measurement data that meets the candidate range in the histogram memory two and store it in the histogram memory one, then initialize the histogram memory two; let i = i + 1, update the next set step size to the current set step size, if i < n, return to step S2; otherwise, go to step S4;

[0051] In step S3, statistically process in the histogram memory one according to the set step size i to generate a reselected histogram, and adopt the same steps as steps S21 - S23. Specifically,

[0052] S31. Set the length of the set step size i to k, and determine the division range as 2 N / 2 k, divide the collected measurement data into 2 k groups according to the divided ranges;

[0053] S32. Calculate the sum of the number of photon flight times in each group obtained in step S31;

[0054] S33. Establish a re-selection histogram for each group obtained in step S31. The abscissa of the re-selection histogram is the serial number of the group, and the ordinate is the number of photon flight times.

[0055] In step S3, determine the candidate range where the peak is located according to the value of the abscissa X i . Specifically, use the group where the value of the abscissa X i corresponding to the second peak is located as the candidate range. The number of measurement data within the candidate range is 2 N / 2 k , and update N = N - k.

[0056] S4. Generate a final selection histogram with a set step size i and perform final selection peak extraction; statistically analyze the measurement data in histogram memory two obtained in step S2 according to the set step size i in histogram memory one, or statistically analyze the measurement data in histogram memory one obtained in step S3 according to the set step size i in histogram memory two to generate a final selection histogram, obtain the final selection peak with the most number of photon flight times in the final selection histogram, store the value of the abscissa Xi corresponding to the final selection peak at the PEAKi address in the PEAK memory, and splice all the abscissa values stored in the PEAK memory to obtain the final photon flight time measurement value.

[0057] In step S4, generate a final selection histogram with a set step size i, and adopt the same steps as steps S21 - S23.

[0058] In step S4, splice the abscissa values stored in the PEAK memory to obtain the final photon flight time measurement value. Specifically, after converting the lengths of the set step sizes corresponding to the abscissa values at the PEAK1 address, PEAK2 address... PEAKi address into binary data, arrange and splice them according to the PEAK1 address, PEAK2 address... PEAKi address, and then the binary value of the final photon flight time measurement value can be obtained.

[0059] This multi-step histogram peak extraction method applied to laser ranging can change the set step size, which is equivalent to adjusting the width of the histogram time interval (the time interval is defined as bin, which is the minimum time accuracy of the histogram), dynamically shrink the statistical range of the histogram, screen out interference data at the fastest speed, directly improve the histogram peak extraction speed, reduce the histogram peak extraction time, and ensure the data accuracy at the same time.

[0060] This multi-step histogram peak extraction method applied to laser ranging reduces the amount of data extraction through noise threshold filtering. The extraction method of variable step size and length greatly reduces the histogram peak extraction time and speeds up the histogram peak extraction speed. Continuously narrowing the extraction range does not cause damage to the valid data, ensuring the accuracy of histogram peak extraction. By multiplexing histogram memory one and histogram memory two under strict timing, the utilization efficiency of hardware resources is improved.

[0061] In this multi-step histogram peak extraction method applied to laser ranging, in steps S2 and S3, initializing histogram memory one and initializing histogram memory two are reset operations for the memory, clearing the internal data of the memory.

[0062] In this multi-step histogram peak extraction method applied to laser ranging, the bit width N of the measurement data is divided into n set step lengths, that is, the sum of the lengths of the n set step lengths is equal to the bit width N of the time-to-digital conversion sensor. The length of each set step length is essentially determined by the data bit width of the time-to-digital conversion sensor. If the data bit width is N, at most N-step histogram peak extraction can be performed. The number of set step lengths and the number of bits per step are determined according to the data distribution range participating in the histogram peak extraction, which is not fixed, and also reflects the flexibility of the method of the present invention. The number of set step lengths and the number of bits per step are determined by the distribution of the noise data filtered by the filter. The distribution includes the number of filtered data and the distribution range of their corresponding abscissas. That is, if the filtered data is more and concentrated in distribution, the number of distributions can be increased, and the corresponding step length will be shortened, and vice versa, to ensure the maximization of peak extraction efficiency. At the same time, the filtering step greatly improves the anti-interference ability of the system.

[0063] A specific example of this multi-step histogram peak extraction method applied to laser ranging is described as follows. For example, Figure 2 , the measurement data collected by the time-to-digital conversion sensor consists of 16-bit binary numbers. There are 65,536 possible abscissas. The black area of the histogram represents the histogram peak, and the white area is the histogram sub-peak or noise. For example, Figure 3 , the 16-bit collected data is divided into three set step lengths, which are 6 bits, 5 bits, and 5 bits respectively.

[0064] First, as Figure 4 (a) shows, the photon flight times of all collected measurement data are statistically sorted into 2 6 = 64 groups according to the number of bits k = 6 of the set step length. The bit width N of the measurement data is 16, so the number of data in each group is 2 16 / 2 6 = 1024. After histogram peak extraction, the value of the abscissa corresponding to the first peak is obtained and stored in the PEAK memory.

[0065] Secondly, as shown in Figure 4 (b), the group where the abscissa corresponding to the first peak is located is used as the candidate range, and the measurement data within the candidate range is selected. The number of measurement data within the candidate range is 2 16-6 = 1024, N = N - k, with the number of bits of the next set step size being 5. After orderly counting in 2 5 = 32 groups, after generating the candidate histogram, the second peak extraction is performed. After the histogram peak extraction, the value of the abscissa corresponding to the first peak is obtained and stored in the PEAK memory.

[0066] Finally, as shown in Figure 4 (c), the group where the abscissa corresponding to the second peak is located is used as the candidate range, and the measurement data within the candidate range is selected. The number of measurement data within the candidate range is 2 10-5 = 32, N = N - k, with the number of bits of the next set step size being 5. After orderly counting in 2 5 = 32 groups, after generating the candidate histogram, the second peak extraction is performed. After the histogram peak extraction, the value of the abscissa corresponding to the first peak is obtained and stored in the PEAK memory. The values of all the abscissas stored in the PEAK memory are concatenated to obtain the final photon flight time measurement value. Specifically, after converting the abscissa values at the PEAK1 address, PEAK2 address... PEAKi address into binary data according to the length of the corresponding set step size, and arranging and concatenating them in the order of the PEAK1 address, PEAK2 address... PEAKi address, the binary value of the final photon flight time measurement value can be obtained.

[0067] The experimental simulation verification of this multi-step histogram peak extraction method applied to laser ranging in the embodiment is as follows: Figure 5 are the main timing waveform diagrams of the peak extraction results this time. They are the clock signal (sys_clk), reset signal (sys_rst_n), peak extraction enable signal (star_en), first step peak (PEAK1), second step peak (PEAK2), third step peak (PEAK3), finally extracted peak (PEAK), and the number of times the peak appears (Num). This simulation test is based on the integrated design environment (vivado) of Xilinx. 1000 16-bit binary value measurement data collected are stored in the memory, and peak extraction is performed. The peak extraction module is enabled to work by the high level of star_en. After the noise filtering circuit works, it is divided into three set step sizes, which are the first set step size with 6 high bits, the second set step size with 5 middle bits, and the third set step size with 5 low bits. After the first peak extraction, the abscissa corresponding to the peak is obtained, as shown in Figure 4(a), According to the set step size 1 with a length of 6, it is converted to a 6-bit binary number 000001; after the second peak extraction, the abscissa corresponding to the peak is obtained, such as Figure 4 (b), According to the set step size 2 with a length of 5, it is converted to a 5-bit corresponding binary 00000; after the third peak extraction, the abscissa corresponding to the peak is obtained, such as Figure 4 (c), According to the set step size 3 with a length of 5, it is converted to a 5-bit corresponding binary 00101; thus, the binary values of the abscissas obtained three times are concatenated in sequence, and the binary value of the final photon flight time measurement value is 0000010000000101 (decimal 1029). At the same time, the number of times Num it appears can be counted as 0000010000 (decimal 16). For a time digital conversion sensor with a measurement range from 0 to 2056 nanoseconds, the photon flight time measurement value can be calculated by 2056 * 1029 / 65536.

[0068] In this multi-step histogram peak extraction method applied to laser ranging, the data bit width of the time digital conversion sensor determines the maximum number of times the histogram peak can be extracted. Under the condition that the distribution times do not exceed the data bit width, three-step, four-step, five-step, etc. histogram peak extraction can be adopted. For example, 16-bit data can be divided into high 4 bits, middle 5 bits, and low 7 bits for three-step histogram peak extraction. Similarly, under the condition that the sum of all step sizes is 16 bits of the data bit width, the data can also be divided into combinations such as high 5 bits, middle 3 bits, and low 8 bits, depending on the data distribution range.

[0069] From this specific example, it can be seen that by setting the noise threshold filtering, the amount of data extraction can be reduced; according to the distribution characteristics of the filtered noise data, the method of dynamically adjusting the number of step sizes and step lengths greatly reduces the histogram peak comparison time, speeds up the histogram peak extraction speed, and improves the anti-interference ability of the system. Continuously narrowing the extraction range does not cause damage to the valid data, ensuring the accuracy of the histogram peak extraction. The reuse of the storage unit improves the utilization efficiency of hardware resources.

[0070] Such as Figure 6 , The embodiment also provides a three-step histogram peak extraction system applied to laser ranging for implementing the three-step histogram peak extraction method applied to laser ranging described in any one of the above. It includes a time digital conversion sensor and a microcontroller unit MCU. The microcontroller unit MCU includes a control module, a noise filtering module, a peak extraction module, a histogram memory one, a histogram memory two, and a peak memory.

[0071] The microcontroller unit MCU enables the time-to-digital conversion sensor, and the time-to-digital conversion sensor starts to collect data and sends the collected data to the microcontroller unit MCU. In the microcontroller unit MCU, after the noise filtering module filters the noise, the measured data is obtained, and the measured data is sent to the control module. The control module divides the bit width N of the measured data into several variable set step lengths, performs dynamic reduction of the peak extraction range processing on the measured data, and alternately performs peak extraction multiple times through the peak extraction module, histogram memory one, and histogram memory two to achieve histogram peak extraction of each set step length after dynamic reduction of the peak extraction range. The abscissas corresponding to the extracted peaks are stored in the peak memory, and the values of all the abscissas stored in the peak memory are concatenated to obtain the final photon flight time measurement value.

[0072] For the multi-step histogram peak extraction method and system applied to laser ranging, the steps of laser ranging include data acquisition and histogram peak extraction. In data acquisition, since the device sensitivities of the laser detectors in the data acquisition part are very high, the detectors are easily affected by noise such as dark counts during detection. Before storing the sensor measurement value, it is compared with the noise threshold. This step filters the measurement value including noise, reduces the data for histogram statistics, reduces the number of times the MCU processes data, and has the effect of shortening the histogram peak extraction time. At the same time, according to the distribution of the filtered noise data, the corresponding number of steps and the length of each step of the variable step length are divided. In histogram peak extraction, due to the data aggregation of the laser ranging histogram, unequal division of the number of bits of the measurement value for histogram extraction can quickly lock the area where the peak is located, minimize the amount of extracted data, and indirectly reduce the number of extractions by the MCU, shortening the extraction time.

[0073] For the multi-step histogram peak extraction method and system applied to laser ranging, the number of step lengths and the step length are flexibly set according to the distance of the measured distance. Essentially, it is to continuously contract the time interval statistically in the histogram with different step lengths until the smallest time interval where the histogram peak is located is found, which can improve the extraction accuracy and avoid the problem that the extraction accuracy is guaranteed during the histogram peak extraction process.

[0074] For the multi-step histogram peak extraction method and system applied to laser ranging, the values that meet the conditions are continuously screened, and the data volume is reduced at the fastest speed, reducing the requirements for hardware conditions. Both histogram memory one and histogram memory two use the internal resources of the MCU, and external memories are not called during data processing, reducing the volume. At the same time, the memory resources are mutually reused, improving the utilization efficiency of the memory.

[0075] The above are only the preferred embodiments of the present invention, and do not impose any form of limitation on the present invention. Although the present invention has been described above with reference to the preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to equivalent embodiments by using the disclosed technical content within the scope of the technical solution of the present invention. However, as long as it does not depart from the technical content of the technical solution of the present invention, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A multi-step histogram peak extraction method applied to laser ranging, characterized in that, Including: Receiving the acquisition data of the time-to-digital conversion sensor and filtering it to obtain measurement data; Based on the quantity and distribution of the measurement data after noise filtering, dividing the bit width N of the measurement data into several variable set steps, and performing dynamic reduction of the peak extraction range on the measurement data. Specifically, generating a histogram with the set step, obtaining the peak in the histogram, storing the abscissa corresponding to the peak, determining the candidate range from the abscissa corresponding to the peak, selecting the measurement data that meets the candidate range, and updating the next set step to the current set step; repeating the dynamic reduction of the peak extraction range process until the last set step generates the final selection histogram, extracting the final peak, storing the abscissa corresponding to the final peak, and splicing the values of all stored abscissas to obtain the final photon flight time measurement value; Specifically including the following steps: S1. Receiving the acquisition data of the time-to-digital conversion sensor and filtering it, extracting the acquisition data higher than the preset noise threshold to obtain measurement data, including the photon flight time and the number of photon flights, and storing them in histogram memory one. At the same time, according to the quantity and distribution of the noise-filtered data, dividing the bit width N of the measurement data into n variable set steps (1…i…n) from high to low successively; S2. Dynamically narrow the peak extraction range of the measurement data, generate a candidate histogram with a set step size i, and perform candidate peak extraction; count the measurement data in the first histogram memory into the second histogram memory according to the set step size i, generate a candidate histogram in the second histogram memory, obtain the first peak with the most photon flight times in the candidate histogram, and store the value of the abscissa X corresponding to the first peak at the PEAKi address of the peak memory. After that, initialize the second histogram memory, determine the candidate range where the peak is located from the value of the abscissa X, select the measurement data in the first histogram memory that meets the candidate range and store it in the second histogram memory, and then initialize the first histogram memory; let i = i + 1, update the next set step size to the current set step size. If i < n, go to the next step S3; otherwise, go to step S4; i After storing the value in the PEAKi address of the peak memory, initialize the second histogram memory. From the value of the abscissa X i determine the candidate range where the peak is located, select the measurement data in the first histogram memory that meets the candidate range and store it in the second histogram memory, and then initialize the first histogram memory; let i = i + 1, update the next set step size to the current set step size. If i < n, enter the next step S3; otherwise, enter step S4; S3. For the measurement data in the histogram memory two obtained in step S2, statistically count them in the histogram memory one according to the set step size i to generate a reselected histogram, obtain the second peak value of the most occurring photon flight times in the reselected histogram, and store the value of the abscissa X corresponding to the second peak value at the PEAKi address in the PEAK memory, and then initialize the histogram memory one; determine the candidate range where the peak is located from the value of the abscissa X, select the measurement data in the histogram memory two that meets the candidate range and store it in the histogram memory one, and then initialize the histogram memory two; let i = i + 1, update the next set step size to the current set step size, if i < n, return to step S2; otherwise, enter step S4; i Store the value at the PEAK memory's PEAKi address, and then initialize the histogram memory one; determine the candidate range where the peak is located from the value of the abscissa X i After selecting the measurement data in the histogram memory two that meets the candidate range and storing it in the histogram memory one, initialize the histogram memory two; let i = i + 1, update the next set step size to the current set step size, if i < n, return to step S2; otherwise, enter step S4; S4. Generating a final selection histogram with the set step i and performing final selection peak extraction; statistically calculating the measurement data in histogram memory two obtained in step S2 in histogram memory one according to the set step i, or statistically calculating the measurement data in histogram memory one obtained in step S3 in histogram memory two according to the set step i to generate the final selection histogram, obtaining the final selection peak with the most photon flight times in the final selection histogram, storing the value of the abscissa Xi corresponding to the final selection peak at the PEAKi address in the PEAK memory, and splicing the values of all stored abscissas in the PEAK memory to obtain the final photon flight time measurement value.

2. The multi-step histogram peak extraction method applied to laser ranging according to claim 1, characterized in that: In step S2, generating a candidate histogram with the set step i, specifically: In step S2, generating a candidate histogram with the set step i, specifically: S21. Set the length of step i to k, and determine that the division range is 2 N / 2 k , and divide the photon flight time of the collected measurement data into 2 k groups according to the division range; S22. Calculating the sum of the number of photon flights in each group obtained in step S21; S23. Establishing a candidate histogram for each group obtained in step S21, where the abscissa of the candidate histogram is the group number and the ordinate is the number of photon flights.

3. The multi-step histogram peak extraction method applied to laser ranging according to claim 2, characterized in that: In step S2, the candidate range where the peak is located is determined by the value of the abscissa X i Specifically, the group where the value of the abscissa X i corresponding to the first peak is located is used as the candidate range, and the number of measurement data within the candidate range is 2 N / 2 k , and update N = N - k.

4. The multi-step histogram peak extraction method applied to laser ranging according to any one of claims 1-3, characterized in that: In step S4, splicing the abscissa values stored in the PEAK memory to obtain the final photon flight time measurement value. Specifically, converting the lengths of the set steps corresponding to the abscissa values of each abscissa in the PEAK1 address, PEAK2 address…PEAKi address into binary data, and then arranging and splicing them according to the PEAK1 address, PEAK2 address…PEAKi address to obtain the binary value of the final photon flight time measurement value.

5. The multi-step histogram peak extraction method applied to laser ranging according to any one of claims 1-3, characterized in that: In step S1, the variable set steps can be either unequal or equal for each set step.

6. A three-step histogram peak extraction system for laser ranging, which implements the three-step histogram peak extraction method for laser ranging described in any one of claims 1-5, characterized in that: Including a time-to-digital conversion sensor and a microcontroller unit MCU. The microcontroller unit MCU includes a control module, a noise filtering module, a peak extraction module, a histogram memory one, a histogram memory two, and a peak memory. The microcontroller unit (MCU) enables the time-to-digital conversion sensor. The time-to-digital conversion sensor starts to collect data and sends the collected data to the microcontroller unit (MCU). In the microcontroller unit (MCU), after the noise filtering module filters the noise, the measured data is obtained and sent to the control module. The control module divides the bit width N of the measured data into several variable set step lengths, performs dynamic peak extraction range reduction processing on the measured data, and alternately performs peak extraction multiple times through the peak extraction module, histogram memory one, and histogram memory two to achieve histogram peak extraction of each set step length after dynamic peak extraction range reduction. The abscissas corresponding to the extracted peaks are stored in the peak memory, and the values of all abscissas stored in the peak memory are concatenated to obtain the final measured value of the photon flight time.

Images