CN107607931B - Laser radar echo image processing method - Google Patents
Laser radar echo image processing method Download PDFInfo
- Publication number
- CN107607931B CN107607931B CN201710664927.9A CN201710664927A CN107607931B CN 107607931 B CN107607931 B CN 107607931B CN 201710664927 A CN201710664927 A CN 201710664927A CN 107607931 B CN107607931 B CN 107607931B
- Authority
- CN
- China
- Prior art keywords
- echo
- original
- oscillogram
- matrix
- processing method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention relates to the field of image processing, in particular to a laser radar echo image processing method, which comprises the following steps: reading in an original echo oscillogram; defining a convolution matrix and a threshold value; performing convolution operation on the convolution matrix and the original echo oscillogram; comparing the operation result with the threshold value to obtain a new matrix with the same size as the original echo oscillogram; multiplying the new matrix with the original echo oscillogram to obtain a middle processing oscillogram; extracting the medium processing oscillogram by using a mathematical average algorithm to obtain space target laser detection echo data; and performing image restoration on the echo data to obtain a target image. The invention can make the echo image clearer.
Description
Technical Field
The invention relates to the field of image processing, in particular to a laser radar echo image processing method.
Background
Laser radar has been rapidly applied to remote measurement, terrain detection, unmanned obstacle avoidance, and the like, despite its short development time, with its unique advantages over conventional measurement techniques. The fringe principle airborne laser radar is taken as a typical representative of the laser radar, and plays an important role in the aspects of target reconnaissance, topographic mapping, underwater target imaging and the like.
At present, in the aspect of hardware, the stripe principle laser radar has already realized the rapid development; in terms of software, the development of hardware is far from pace. Especially, the preprocessing part of the original echo signal has large calculation amount and higher difficulty, and becomes an important factor for restricting the overall development of the laser radar at present.
One of the prior art solutions is to set the intensity of the fringe echo signal, and directly extract the portion with intensity exceeding the set value. The other scheme is that a special function is used for fitting the fringe signal, and fitting function information is extracted to serve as laser echo information.
Although the direct set value extraction method is convenient to implement, a preset value needs to be selected manually, and the selection of the preset value has a decisive influence on the signal extraction effect. In addition, the difference of the optimal preset values under different working conditions in the actual test process is very large, and a professional with rich experience is required to extract signals. The preset value extraction method has the problems of low extraction precision, manual participation and the like.
The function fitting method has the problems of large operation amount, low extraction efficiency and the like. And an echo signal function model needs to be obtained in advance, so that the automatic extraction of the signal is not facilitated.
Therefore, it is urgent to develop a method for accurately extracting target image information from echo signals.
Disclosure of Invention
The embodiment of the invention provides a laser radar echo image processing method, which aims to solve the technical problem that the existing target echo image is not clear in extraction.
The laser radar echo image processing method provided by the embodiment of the invention comprises the following steps:
step S101: reading in an original echo oscillogram;
step S102: defining a convolution matrix and a threshold value;
wherein the convolution matrix
Wherein a isij1, (i is 1, …, N, j is 3,5,7,9), N is a natural number.
Step S103: performing convolution operation on the convolution matrix and the original echo oscillogram; comparing the operation result with the threshold value to obtain a new matrix with the same size as the original echo oscillogram;
step S104: multiplying the new matrix with the original echo oscillogram to obtain a middle processing oscillogram;
step S105: extracting the medium processing oscillogram by using a mathematical average algorithm to obtain space target laser detection echo data;
step S106: and performing image restoration on the echo data to obtain a target image.
Further, in step S102: n is 1/20 for a single row of pixels of the original echo waveform map.
Further, wherein the convolution matrix
Wherein a isij＝1，(i＝1,…,50；j＝3,5,7,9)。
Further, in step S103, the step of comparing the operation result with the threshold to obtain a new matrix having the same size as the original echo waveform map specifically includes: the more than threshold part is directly assigned 1 at the corresponding pixel position, and the less than threshold part is directly assigned 0 at the corresponding pixel position, thereby obtaining a new matrix of the same size as the original echo waveform map.
Further, the threshold value is set to 1 to 100.
Further, the threshold value is set to 1 to 10.
Further, the threshold value is set to 1.
Further, the step 104 multiplies as follows:
p(x，y)＝f(x，y)g(m，n)＝p(xm，yn) (1-2)
wherein f (x, y) is the gray value in the original image; h (x, y) represents a normalized sliding window; (M, N) is the size of the original fringe pattern; p (x, y) represents the gray value of the denoised image; g (m, n) is the gray value of the new matrix herein.
Assuming O as the original echo waveform map, A as the new matrix and B as the intermediate processing matrix, the process can be expressed as a matrix convolution
B＝O*A
Further, the mathematical averaging algorithm in step 105 is as follows:
wherein IijThe gray value of each pixel point on the two-dimensional gray image is obtained.
The invention provides a laser radar echo image processing method, which can perfectly reserve the detail information of the waveform signal edge through an EDT (edge detectetionrecording) technology and can filter the influence of background noise, thereby accurately restoring the information of a target object and ensuring the accuracy of radar detection.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.
FIG. 1 is a flow chart of a method according to one embodiment of the present invention;
FIG. 2 is a graph of the original fringe signal according to one embodiment of the present invention;
FIG. 3 is a graph of the results of a prior art wiener process;
FIG. 4 is a graph of the results of EDT processing according to one embodiment of the present invention;
FIG. 5 is a graph of the size of the different convolution matrices (A) versus correlation coefficients according to the present invention;
FIG. 6 is a graph of different threshold values versus correlation coefficients according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the present invention will be described in further detail with reference to the accompanying drawings, and it is apparent that the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The laser radar original signal can obtain laser point cloud data after information extraction, and the laser point cloud data is a unified data form for later-stage production of various digital electronic map products. The original echo signal of the laser radar is a fringe pattern due to the fringe principle. The abscissa of the bar chart represents time information of an echo signal, that is, distance information of a target; the ordinate of the image represents the spatial position information of the target, that is, the laser diverges in space, when the laser reaches the detection target after being emitted, the laser diverges in space and has a certain divergence angle, and the light rays in different directions irradiate different positions of the target, so that the divergence of the echo signal on the ordinate represents the spatial position information of the detection target. Therefore, the extraction of the stripe echo signal is the extraction of the three-dimensional position information and the surface reflection intensity information of the detection target, so that abundant information of the detection target can be obtained.
Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
Example 1
The method (EDT) comprises the following main processes of firstly carrying out convolution processing and denoising on an original echo signal based on an edge detection theory to obtain a middle processing oscillogram, and then extracting the middle processing oscillogram by utilizing a mathematical average algorithm to obtain space target laser detection echo data. The specific flow for obtaining the waveform diagram of the middle processing is shown in fig. 1.
The laser radar echo image processing method provided by the embodiment of the invention comprises the following steps:
step S101: reading in an original echo oscillogram;
step S102: defining a convolution matrix and a threshold value;
for example, the convolution matrix may be
Wherein a isij1, (i is 1, …, N, j is 3,5,7,9), N is a natural number.
Step S103: performing convolution operation on the convolution matrix and the original echo oscillogram; comparing the operation result with the threshold value to obtain a new matrix with the same size as the original echo oscillogram;
step S104: multiplying the new matrix with the original echo oscillogram to obtain a middle processing oscillogram;
step S105: extracting the medium processing oscillogram by using a mathematical average algorithm to obtain space target laser detection echo data;
step S106: and performing image restoration on the echo data to obtain a target image.
Further, in step S101, an echo waveform map may be obtained according to the received laser detection echo scan map, and the echo map data may be stored, and after removing noise (removing noise in a laser point cloud, a point below the ground, a cloud, or a bird in flight, etc.), preliminary graphic data may be obtained, so as to obtain processable data.
Further, for the step S102: the definition of the convolution matrix generally needs to be determined according to the pixels of the original echo waveform map, and N can be selected as 1/20 of a single row of pixels of the original echo waveform map. For example, when the original echo waveform pixels are 1000 × 1000, N is 50, and the matrix is 50 × 50, as shown below. Of course N may also be 1/10 or 1/30 for a single row of pixels of the original echo waveform map. Through experimental simulation, the value selection 1/20 has a good technical effect, and a clear target image can be obtained. The convolution matrix is as follows:
convolution matrix
Wherein a isij＝1，(i＝1,…,50；j＝3,5,7,9)。
Further, in step S103, the step of comparing the operation result with the threshold to obtain a new matrix having the same size as the original echo waveform map specifically includes: for the part larger than the threshold value, 1 is directly assigned at the corresponding pixel position, and for the part smaller than the threshold value, 0 is directly assigned at the corresponding pixel position, so that a new matrix with the same size as the original echo waveform map is obtained, as shown in the following matrix.
And multiplying the new matrix and the original echo oscillogram to obtain a middle processing oscillogram.
In the above steps, the smaller the threshold value is, the better the threshold value is, the smaller the value is, the higher the definition of the obtained graph is through the comparison operation, but the requirement on the calculation amount is also higher, therefore, the definition and the calculation amount are balanced to select a suitable threshold value range, for example, the threshold value range can be set to 1-100. Preferably, the threshold is set to 1-10. The most preferable threshold is set to 1, and in this case, the best image clarity can be obtained.
The echo signals obtained under different experimental conditions are processed, so that the output effect of the EDT method is insensitive to the characteristic values such as a matrix form, a threshold value and the like, and the automatic waveform signal extraction is favorably realized. As shown in fig. 5 and 6.
Step S104: multiplying the new matrix with the original echo oscillogram to obtain a middle processing oscillogram;
step S105: extracting the medium processing oscillogram by using a mathematical average algorithm to obtain space target laser detection echo data;
wherein, the multiplication operation of step 104 is as follows:
p(x，y)＝f(x，y)g(m，n)＝p(xm，yn) (1-2)
wherein f (x, y) is the gray value in the original image; h (x, y) represents a normalized sliding window; (M, N) is the size of the original fringe pattern; p (x, y) represents the gray value of the denoised image; g (m, n) is the gray value of the new matrix herein.
Let O be the original echo waveform diagram (consider the original echo waveform diagram as a matrix of a certain size), A be the new matrix, and B be the middle processing matrix (waveform diagram). The process can be represented as a matrix convolution
B＝O*A
Wherein, the mathematical averaging algorithm in step 105 is as follows:
wherein IijThe gray value of each pixel point on the two-dimensional gray image is obtained.
The embodiment of the invention provides a laser radar echo image processing method, which can perfectly reserve the edge detail information of a waveform signal and filter the influence of background noise by an EDT (edge detectetionrecording) technology, thereby accurately restoring the information of a target object and ensuring the accuracy of radar detection.
The present invention applies a graphical means to the waveform signal extraction process. The method can obviously improve the processing speed of mass laser echo signals and has proved to have better signal extraction effect.
The method (EDT) provided by the invention can perfectly reserve the detail information of the waveform signal edge and can filter the influence of background noise. The comparison of the intermediate processing waveforms obtained by the method provided by the invention and the conventional wiener processing method is as follows: fig. 2 is a graph of an original fringe signal, fig. 3 is a graph of a result obtained by a wiener processing method, and fig. 4 is a graph of a processing result of an EDT method according to the present invention.
The quantitative statistics obtained for 3000 fringe signals are shown in the following table (SNR is signal-to-noise ratio and r is correlation coefficient). Therefore, the EDT can effectively improve the signal-to-noise ratio of the image and can also ensure that the correlation coefficient of the processed image and the original signal is higher.
The above-described apparatus embodiments are merely illustrative, wherein the units described as separate components may or may not be physically separate. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (8)
1. A laser radar echo image processing method comprises the following steps:
step S101: reading in an original echo oscillogram;
step S102: defining a convolution matrix and a threshold value;
wherein the convolution matrix
Wherein a isij1, …, N; j is 3,5,7,9, and N is a natural number;
step S103: performing convolution operation on the convolution matrix and the original echo oscillogram; comparing the operation result with the threshold value to obtain a new matrix with the same size as the original echo oscillogram;
step S104: multiplying the new matrix with the original echo oscillogram to obtain a middle processing oscillogram;
step S105: extracting the medium processing oscillogram by using a mathematical average algorithm to obtain space target laser detection echo data;
step S106: and performing image restoration on the echo data to obtain a target image.
2. The lidar echo image processing method according to claim 1, wherein in step S102: n is 1/20 for a single row of pixels of the original echo waveform map.
3. The lidar echo image processing method according to claim 2,
wherein the convolution matrix
Wherein a isij＝1，i＝1,…,50；j＝3,5,7,9。
4. The lidar echo image processing method according to claim 1, wherein in step S103, the operation result is compared with the threshold, and a new matrix having the same size as the original echo waveform map is obtained by: the more than threshold part is directly assigned 1 at the corresponding pixel position, and the less than threshold part is directly assigned 0 at the corresponding pixel position, thereby obtaining a new matrix of the same size as the original echo waveform map.
5. The lidar echo image processing method according to claim 4, wherein the threshold value is set to 1 to 100.
6. The lidar echo image processing method according to claim 5, wherein the threshold value is set to 1 to 10.
7. The lidar echo image processing method according to claim 6, wherein the threshold value is set to 1.
8. The lidar echo image processing method according to claim 1, wherein the step S104 is multiplied as follows:
p(x，y)＝f(x，y)g(m，n)＝p(xm，yn)
wherein f (x, y) is the gray value in the original image; h (x, y) represents a normalized sliding window; (M, N) is the size of the original fringe pattern; p (x, y) represents the gray value of the denoised image; g (m, n) is the gray value of the new matrix;
assuming O as the original echo waveform map, A as the new matrix and B as the intermediate processing matrix, the process can be expressed as a matrix convolution
B＝O*A
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710664927.9A CN107607931B (en) | 2017-08-07 | 2017-08-07 | Laser radar echo image processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710664927.9A CN107607931B (en) | 2017-08-07 | 2017-08-07 | Laser radar echo image processing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107607931A CN107607931A (en) | 2018-01-19 |
| CN107607931B true CN107607931B (en) | 2020-06-05 |
Family
ID=61064362
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710664927.9A Active CN107607931B (en) | 2017-08-07 | 2017-08-07 | Laser radar echo image processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107607931B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118151126A (en) * | 2022-12-07 | 2024-06-07 | 上海禾赛科技有限公司 | Laser radar, data processing method and light detection and data acquisition processing device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101839981A (en) * | 2010-01-28 | 2010-09-22 | 中国人民解放军电子工程学院 | Method and device for acquiring laser imaging echo waveform and level characteristics |
| CN101963666A (en) * | 2010-09-20 | 2011-02-02 | 北京理工大学 | Method for correcting intensity image distortion of streak tube imaging laser radar |
| CN105572649A (en) * | 2015-12-11 | 2016-05-11 | 中北大学 | Radar target detection method based on sparse Fourier transform |
| US9598076B1 (en) * | 2015-10-22 | 2017-03-21 | Ford Global Technologies, Llc | Detection of lane-splitting motorcycles |
| CN106599808A (en) * | 2016-12-01 | 2017-04-26 | 中国科学院光电研究院 | Full-waveform laser radar data based concealed target extraction method |
-
2017
- 2017-08-07 CN CN201710664927.9A patent/CN107607931B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101839981A (en) * | 2010-01-28 | 2010-09-22 | 中国人民解放军电子工程学院 | Method and device for acquiring laser imaging echo waveform and level characteristics |
| CN101963666A (en) * | 2010-09-20 | 2011-02-02 | 北京理工大学 | Method for correcting intensity image distortion of streak tube imaging laser radar |
| US9598076B1 (en) * | 2015-10-22 | 2017-03-21 | Ford Global Technologies, Llc | Detection of lane-splitting motorcycles |
| CN105572649A (en) * | 2015-12-11 | 2016-05-11 | 中北大学 | Radar target detection method based on sparse Fourier transform |
| CN106599808A (en) * | 2016-12-01 | 2017-04-26 | 中国科学院光电研究院 | Full-waveform laser radar data based concealed target extraction method |
Non-Patent Citations (2)
| Title |
|---|
| Study on One-Dimensional Scanning Semiconductor Lidar;Zhao-shuo Tian et a.;《Innovation for Applied Science and Technology》;20130125;全文 * |
| 条纹管成像激光雷达条纹数据提取算法;马晨宁 等;《中国激光》;20120430;第39卷(第4期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107607931A (en) | 2018-01-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6057881B2 (en) | Method for removing noise from input image consisting of pixels containing noise | |
| Di Martino et al. | Benchmarking framework for SAR despeckling | |
| CN109583378A (en) | A kind of vegetation coverage extracting method and system | |
| Paris et al. | A three-dimensional model-based approach to the estimation of the tree top height by fusing low-density LiDAR data and very high resolution optical images | |
| EP3480782A1 (en) | Method and device for reducing noise in a depth image | |
| US9651661B2 (en) | Methods and systems for local principal axis rotation angle transform | |
| US9183671B2 (en) | Method for accelerating Monte Carlo renders | |
| DE102016106511A1 (en) | Parametric online calibration and compensation for TOF imaging | |
| CN104200434B (en) | Non-local mean image denoising method based on noise variance estimation | |
| CN104217406A (en) | SAR image noise reduction method based on shear wave coefficient processing | |
| CN103871031A (en) | Kernel regression-based SAR image coherent speckle restraining method | |
| CN109785242A (en) | A kind of solution mixing method based on the high spectrum image by wave band generalized bilinear model | |
| CN117115669B (en) | Object-level ground object sample self-adaptive generation method and system with double-condition quality constraint | |
| Fu et al. | An anisotropic Gaussian filtering model for image de-hazing | |
| CN107607931B (en) | Laser radar echo image processing method | |
| Sebastianelli et al. | A speckle filter for Sentinel-1 SAR ground range detected data based on residual convolutional neural networks | |
| Bolsee et al. | CNN-based denoising of time-of-flight depth images | |
| CN107835998A (en) | For identifying the layering Tiling methods of the surface type in digital picture | |
| CN116823913A (en) | Non-visual field image reconstruction method, computer readable storage medium and terminal equipment | |
| Mu et al. | An Enhanced High-Order Variational Model Based on Speckle Noise Removal With $ G^ 0$ Distribution | |
| CN107607927B (en) | Stripe laser echo information extraction method | |
| CN115601301B (en) | Fish phenotype characteristic measurement method, system, electronic equipment and storage medium | |
| CN116740490A (en) | Side-scan sonar image sample amplification and target detection method based on sonar equation | |
| Santhosh et al. | A novel approach for denoising coloured remote sensing image using Legendre Fenchel transformation | |
| CN112465697A (en) | Offshore foggy day image simulation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |