CN103033857A - Rainfall and snowfall automatic observation method based on parallel light large visual field - Google Patents

Abstract

Provided is a rainfall and snowfall automatic observation method based on a parallel light large visual field. Data is transmitted to a terminal computer and corresponding software modules through utilization of an optical unit, a linear array imaging sensor array, a data acquisition control unit and a transmission unit. The optical unit comprises a large visual field collimator and an imaging system. The collimator comprises a collimated light source, a diaphragm and a collimating lens set. A high-brightness light emitting diode (LED) light source or a semiconductor laser is adopted as the collimated light source, and even and stable parallel light is obtained through a piece of ground glass, the diaphragm and the collimating lens set. An imaging camera lens is arranged in front of the outgoing end of the collimator, receives the parallel light emitted from the collimator, and projects the parallel light on a linear array imaging sensor after the parallel light is optically enlarged. The shape, the equivalent size, the falling end speed and other information of a single rainfall grain can be obtained, and therefore automatic identification of rainfall phenomenon and calculation of statistical characteristics of rainfall micro-physical parameters can be achieved.

Description

Precipitation snowfall automatic observation process based on parallel wide visual field

Technical field

The present invention relates to a kind of meteorological observation method of surface Weather phenomenon automatic Observation, particularly a kind of method that is applicable to precipitation snowfall phenomenon is carried out automatic measurement.

Background technology

That precipitation snowfall phenomenon refers to is liquid, water-setting solid-state or mixed state (freezing) thing is from dropping to ground a kind of weather phenomenon in the air.This type of weather phenomenon and human lives are closely bound up, can cause drought and waterlogging such as the inequality of Precipitation Distribution in Time and Space, and in short-term precipitation may cause mud-stone flow disaster in some region; Continuous strong snowfall meeting causes the paralysis of traffic large tracts of land, causes agricultural poor harvest or Severe Reduction.In water globe thermal cycle and weather Changeement, precipitation data is one of main meteorological data.In addition, the precipitation measurement data can be used for the fields such as weather phenomenon is identified automatically, the research of cloud and mist microphysical processes, radar echo intensity verification.Therefore, the quantitative observation of realizing precipitation snowfall phenomenon is for atmospheric science research and prevent and reduce natural disasters and have great significance.

The observation of precipitation snowfall phenomenon only limits to utilize rain gage bucket in the current operation, tilting bucket rains etc. obtain the macroscopic physical quantities such as precipitation intensity and quantity of precipitation, its spatial and temporal resolution is all lower, and the identification of type of precipitation depends on the artificial visually examine, and can't obtain the phase of precipitation particles, shape, the microscopic information such as speed and particle spectra, be not specifically designed to the instrument of snowfall observation in the business, therefore the observation of current precipitation snowfall phenomenon can't be satisfied scientific research, the departments such as production and military affairs are day by day strongly to the more further understanding of precipitation, and the demand of precipitation snowfall information automation and the measurement that becomes more meticulous.

Realize the automatic Observation of precipitation snowfall phenomenon, can change on the one hand the parallel situation of current artificial observation and apparatus measures, help to improve observation quality, satisfy the demand of encrypting observation, for the weather conditions of grasping China, monitoring and early warning to especially big meteorological disaster, for safeguarding China's traffic safety, preventing and reducing natural disasters and Environmental security etc. provides scientific basis, on the other hand, obtain the information such as high-resolution precipitation particles spectrum data and instantaneous precipitation intensity, can be atmospheric science research and the related application demand provides how valuable information.

Optics precipitation phenomenon automatic measurement technology originates in the sixties in 20th century, at present OSI, OWI series, the sensors such as the VPF-730 of Britain Biral, Hungarian PWS100 of LEDWI, the WIVIS of the PWD series that Finland Vaisala company is arranged of comparative maturity, U.S. ASOS.Acquisition technology has its unique advantage in the optical measuring technique, shape and the oscillating characteristic of single precipitation particles can be provided, two-dimensional video precipitation measuring instrument (the Two-Dimensional Video Disdrometer based on the shooting imaging such as developments such as Schonhuber, 2-DVD) and the development such as D.Baumgardner block two-dimensional video precipitation measuring instrument (the Meteorological particle Sensor of image-forming principle based on directional light, MPS), the least resolution of 2DVD is 0.19mm, to the small-particle detectivity a little less than, MPS has improved the detectivity to small-particle, least resolution can reach 0.05mm, but its sampling area only is 200mm*3.2mm, to the detectivity of macroparticle a little less than, the precipitation particles spectrum information of gamut can't be provided.

Aspect practical application, China's part station and scientific research institutions' import a small amount of instrument and equipment, be mainly used in the meteorological support of scientific research and major event, except this quasi-instrument was expensive, whether its performance parameter meets China's region characteristic, job stability and data reliability also remained further to be checked.Therefore, no matter be from service operation or the angle of scientific research, Development of Domestic robotization and the precipitation surveying instrument that becomes more meticulous are current urgent tasks.

Summary of the invention

The present invention seeks to: in the problem that exists in the above-mentioned traffic observation and the current atmospheric science research to precipitation snowfall robotization with the demand that becomes more meticulous and measure, the invention provides a kind of novel precipitation and snowfall phenomenon measuring system and corresponding measuring method, realize the continuous of precipitation phenomenon, automatically unmanned observation, the information such as the shape of single precipitation particles and snowflake and whereabouts end speed can be provided, for further carrying out precipitation intensity, the automatic identification of precipitation character and type of precipitation provides condition, for the automatic observation that promotes weather phenomenon, further promote the full key element automatic observation of Ground Meteorological important in inhibiting.

For achieving the above object, technical scheme of the present invention is, adopts a kind of precipitation phenomenon automatic observing system of the precipitation snowfall imaging based on parallel wide visual field, and this system comprises optical unit, line scan image sensor, data acquisition control unit and transmission unit; Transfer to terminal computer and corresponding software module.Wherein, described optical unit (measurement module) comprises parallel light tube and imaging system two parts.Parallel light tube comprises source of parallel light, diaphragm and collimation lens set, and source of parallel light adopts high-brightness LED light source or semiconductor laser, obtains even, stable directional light through after frosted glass, diaphragm and the collimation lens set; Imaging lens is positioned at the place ahead of parallel light tube exit end, receives the directional light that sends from parallel light tube, projects on the line scan image sensor through optical amplifier.Wherein, line scan image sensor is spliced by two line scan image sensors of identical performance, and its data output and exposure arrange by the unified control in data acquisition control unit.High-brightness LED light source or semiconductor laser obtain even, stable directional light through after frosted glass, diaphragm and the collimation lens set; Parallel light tube when the imaging lens distance and the diameter of parallel light tube determined sampling area, greater than 10cm ²

Described data acquisition control unit and transmission unit are take programmable logic device (PLD) FPGA as core, and the sequential of finishing line array sensor drives, collection and buffer memory, linear array pretreatment and the transmission interface logic of high speed linear array view data.

After imageing sensor places imaging lens, under the control of data acquisition control unit, carry out high-velocity scanning, obtain the image information of precipitation particles; Because the precipitation particles of different scale and the whereabouts end speed of snowflake have a very wide distribution, two line scan image sensors adopt different frequency of exposure, namely obtain precipitation particles by the projection of sampling interval with the frequency of exposure of 25000 frame/seconds, 2500 frames/second frequency of exposure is obtained the snowfall particle and is passed through sampling interval, obtain the two dimensional image of precipitation particles by the picture mosaic to projection, so utilize two dimensional image to extract to obtain the shape of precipitation particles, equivalent yardstick is diameter and end speed information; Based on the precipitation phenomenon automatic observation process of parallel wide visual field, can realize precipitation snowfall phenomenon continuously, robotization and the measurement that becomes more meticulous; Employing comprises that optical unit, line scan image sensor array, data acquisition control unit and transmission unit carry out automatic Observation, optical unit comprises parallel light tube and imaging lens, line scan image sensor places behind the imaging lens under the control of data acquisition control unit, carry out high-velocity scanning, obtain the image information of precipitation particles; Parallel light tube has been determined sampling area to the diameter of the distance between the imaging lens and parallel light tube.Adopt line scan image sensor to obtain precipitation particles by the projection of sampling interval with certain frequency of exposure, obtain the two dimensional image of precipitation particles by the picture mosaic to projection, and then utilize two dimensional image to extract the information such as the shape of precipitation particles, equivalent yardstick (diameter) and end speed that obtain.

Adopt following method to calculate precipitation particles diameter and end speed: the precipitation particles diameter obtains by precipitation particles image analysis in the two dimensional image, or directly calculate: the pixel number that is blocked according to the corresponding moment, pixel dimension and optical imagery enlargement ratio obtain the horizontal scale of corresponding moment precipitation particles, and constantly scanning projection amalgamation of difference can be obtained two dimensional image and obtains the precipitation particles shape information; All projection amalgamations just can be obtained the complete two dimensional image of precipitation particles, according to the valid data frame number of linear array exposure sweep frequency and the precipitation particles that obtains, calculate the whereabouts end speed of precipitation particles:

Be D for diameter _EqPrecipitation particles, its landing end speed V _tFor:

$V_t=\frac{D_{\mathrm{eq}}f}{M}$

F is the CCD frequency of exposure in the formula, and M is that equivalent diameter is D _EqNumber of active lines corresponding to raindrop.

The whereabouts end speed of precipitation particles diameter and precipitation particles can obtain corresponding weather phenomenon type; Being aided with the precipitation particles shape information judges accurately.

Obtain precipitation particles by the projection of sampling interval with the frequency of exposure of 25000 frame/seconds, obtain snowflake by the projection of sampling interval 2500 frame/seconds with lower frequency of exposure, adopt low frequency of exposure to obtain the pattern distortion that the snowflake image can avoid the high speed over-sampling to cause.

Beneficial effect of the present invention is: the signal processing method that adopts line scan image sensor splicing and different frequency of exposure, cooperate lower by above-mentioned hardware and software, this system can realize automatic, the continuous and measurement that becomes more meticulous of precipitation and snowfall phenomenon, the information such as the shape of single precipitation particles and snowflake, equivalent yardstick, whereabouts end speed can be provided, high time resolution precipitation particles spectrum information can be provided on this basis, for the automatic identification that realizes type of precipitation provides the basis, and can submit necessary information for the automatic identification of weather phenomenon.

Description of drawings

Fig. 1 precipitation snowfall phenomenon automatic observing system functional schematic

Fig. 2 precipitation particles imaging synoptic diagram

Fig. 3 flow chart of data processing figure

Embodiment

Below in conjunction with accompanying drawing, describe embodiments of the present invention in detail.

Parallel wide visual field precipitation snowfall phenomenon automatic observing system as shown in Figure 1, described system comprises optical system, data acquisition transport module and terminal computer.Described optical system comprises parallel light tube and imaging system, and source of parallel light provides even, stable directional light, for line scan image sensor provides with reference to light intensity value.Imaging system is incident upon directional light on the line array sensor by certain multiplying power, and sensor gathers with the light intensity variation of certain frequency of exposure to directional light.

As shown in Figure 2, all shades section combination just can be obtained the complete two dimensional image of precipitation particles, expose sweep frequency and the valid data frame number of the precipitation particles that obtains according to linear array, can calculate the whereabouts end speed of precipitation particles.

Be D for equivalent diameter _EqRaindrop, its landing end speed V _tFor:

$V_t=\frac{D_{\mathrm{eq}}f}{M}$

F is the CCD frequency of exposure in the formula, and M is equivalent diameter D _EqNumber of active lines corresponding to raindrop.On the basis of the equivalent diameter that obtains precipitation particles and whereabouts end speed, can calculate relevant Microphysical Characteristics.

Such as Fig. 3, according to the slice information of obtaining, by obtaining the reconstructed image of particle after picture mosaic and the correction, carry out on this basis Image outline identification, extract the features such as yardstick, end speed, axial ratio of particle, then can obtain the information such as yardstick shop, velocity spectrum distribution, type of precipitation of particle by statistical study.

Graphical analysis obtains the equivalent diameter D of precipitation particles _Eq, precipitation intensity can obtain by following formula:

$R=\frac{\mathrm{\π}}{6}\mathrm{\ΣN}\left(D\right)\mathrm{\ρ}{D_{\mathrm{eq}}}^{3}V\left(D\right)$

N in the formula (D) distributes for particle spectra, and V (D) is the sinking speed of raindrop or snowflake, and ρ is the density of raindrop or snowflake, obtains by empirical relationship.

Precipitation snowfall measuring method of the present invention: can obtain the original-shape information of precipitation particles, rely on the later image treatment technology, the microphysical property that can be research precipitation particles and snowflake provides the basis.The linear array precipitation measuring according to the shape of particle information of obtaining, can be realized the detection of rainfall particle phase information in conjunction with temperature, humidity, for the automatic identification of type of precipitation and weather phenomenon provides basis and necessary information.

Claims (8)

1. based on the precipitation snowfall automatic observation process of parallel wide visual field, it is characterized in that utilizing optical unit, line scan image sensor array, data acquisition control unit and transmission unit, transfer to terminal computer and corresponding software module; Wherein, described optical unit comprises large visual field parallel light tube and imaging system two parts; Parallel light tube comprises source of parallel light, diaphragm and collimation lens set, and source of parallel light adopts high-brightness LED light source or semiconductor laser, obtains even, stable directional light through after frosted glass, diaphragm and the collimation lens set; Imaging lens is positioned at the place ahead of parallel light tube exit end, receives the directional light that sends from parallel light tube, projects on the line scan image sensor through optical amplifier; Wherein, line scan image sensor is spliced by two line scan image sensors of identical performance, and its data output and exposure arrange by the unified control in data acquisition control unit; High-brightness LED light source or semiconductor laser obtain even, stable directional light through after frosted glass, diaphragm and the collimation lens set; Parallel light tube when the imaging lens distance and the diameter of parallel light tube has been determined sampling area and greater than 10cm ²Described data acquisition control unit and transmission unit are take programmable logic device (PLD) FPGA as core, and the sequential of finishing line array sensor drives, collection and buffer memory, linear array pretreatment and the transmission interface logic of high speed linear array view data;

Imageing sensor places behind the imaging lens under the control of data acquisition control unit, carries out high-velocity scanning, obtains the image information of precipitation particles; Adopt line scan image sensor to obtain precipitation particles by the projection of sampling interval with the frequency of exposure of 25000 frame/seconds, 2500 frames/second frequency of exposure is obtained the snowfall water particle and is passed through sampling interval, obtain the two dimensional image of precipitation particles by the picture mosaic to projection, so two dimensional image extract obtain the shape of precipitation particles, equivalent yardstick is diameter and end speed information;

Adopt following method to calculate precipitation particles diameter and end speed: the precipitation particles diameter obtains by precipitation particles image analysis in the two dimensional image, or directly calculate: the pixel number that is blocked according to the corresponding moment, pixel dimension and optical imagery enlargement ratio obtain the horizontal scale of corresponding moment precipitation particles, and constantly scanning projection amalgamation of difference is obtained two dimensional image and obtains the precipitation particles shape information; All projection amalgamations just can be obtained the complete two dimensional image of precipitation particles, according to the valid data frame number of linear array exposure sweep frequency and the precipitation particles that obtains, calculate the whereabouts end speed of precipitation particles:

Be D for diameter _EqRaindrop, its landing end speed V _tFor:

$V_t=\frac{D_{\mathrm{eq}}f}{M}$

F is the CCD frequency of exposure in the formula, and M is that equivalent diameter is D _EqNumber of active lines corresponding to raindrop; The whereabouts end speed of precipitation particles diameter and precipitation particles can obtain corresponding weather phenomenon type; Being aided with the precipitation particles shape information judges accurately.

2. according to claim 1 based on the precipitation snowfall phenomenon automatic observation process of parallel wide visual field, it is characterized in that the whereabouts end speed of precipitation particles diameter and precipitation particles can obtain corresponding meteorological condition; Being aided with the precipitation particles shape information judges accurately.

3. according to claim 1 based on the precipitation snowfall phenomenon automatic observation process of parallel wide visual field, it is characterized in that the valid data frame number in conjunction with linear array frequency of exposure and the precipitation particles that obtains, calculate the whereabouts end speed of precipitation particles.

4. according to claim 1 or the 4 precipitation snowfall phenomenon automatic observation process based on parallel wide visual field, it is characterized in that parallel light tube when the imaging lens distance and the diameter of parallel light tube determined sampling area.

5. according to claim 1 based on the precipitation snowfall phenomenon automatic observation process of parallel wide visual field, it is characterized in that adopting two line scan image sensor splicings, two line scan image sensors adopt different exposure rate, obtain the different images feature of synchronization precipitation snowfall phenomenon, in conjunction with weather informations such as humitures, help to carry out the automatic identification of type of precipitation and weather phenomenon.

6. according to claim 1 based on the precipitation snowfall phenomenon automatic observation process of parallel wide visual field,, it is characterized in that adopting low exposure rate to obtain the less snowflake image of whereabouts end speed, can prevent that over-sampling from causing the distortion of image.

7. according to claim 1 based on the precipitation snowfall phenomenon automatic observation process of parallel wide visual field, it is characterized in that adopting two line scan image sensors to obtain the response characteristic of current weather phenomenon with different exposure rate, according to the response characteristic of line scan image sensor under two kinds of conditions of exposures, be applied to the correction of parasitic light inhibition and noise.

8. according to claim 1 based on the precipitation snowfall phenomenon automatic observation process of parallel wide visual field, it is characterized in that obtaining by graphical analysis the equivalent diameter D of precipitation particles _Eq, precipitation intensity obtains by following formula:

$R=\frac{\mathrm{\π}}{6}\mathrm{\ΣN}\left(D\right)\mathrm{\ρ}{D_{\mathrm{eq}}}^{3}V\left(D\right)$

N in the formula (D) distributes for particle spectra, and V (D) is the sinking speed of raindrop or snowflake, and ρ is the density of raindrop or snowflake.

Images