FI113894B - Remote Sensing Method - Google Patents

Description

χ 113894

FIELD OF THE INVENTION FIELD OF THE INVENTION

The invention relates to remote sensing. In particular, the invention relates to the determination of water quality from remote sensing material.

BACKGROUND OF THE INVENTION

The state of nature is currently assessed in many different ways. The purpose of the evaluation is to carry out scientific research and study of the nature load. By examining the load, the cause of the natural pollution can be identified and the load can be changed. Agriculture is one of the biggest polluters of nature. Agricultural fertilizers are leached into waterways where 15 fertilizers cause nutrients such as Rehe. By analyzing the load and pollutants it is possible to reduce the amount of pollutants.

Traditionally, analysis is done by collecting samples. In the case of watercourses, this means taking water samples from several different sampling points. The water samples are analyzed in the laboratory and the results are collected in a database. The results allow conclusions to be drawn and recommendations made, for example, on the fertilizers used. Similar go - '. * ·. The methods can also be used to observe other nature, such as collecting soil samples. The benefit of analyzing samples is the accuracy of the results. All the substances to be examined can be accurately sampled and quantified. Collection of samples ·. The problem with picking is the slowness of collecting. If desired λ. Because of the large number of samples it takes to collect them! . time and work. Therefore, collecting samples is also expensive.

Instead of collecting samples, it is possible to analyze the V 35 by nature. The desired terrain can be photographed from an airplane or satellite.

113894 2

The image captures spectra reflected from the terrain. Such three-dimensional spectral images are known, for example, from International Patent Publication No. WO 01/54067. International patent publication WO 01/54067 5 also describes various methods of sorting and classifying spectral images. Remote sensing of watercourses from an aircraft is described, for example, in U.S. Patent Nos. 4,986,655 and 4,986,656. In particular, they deal with the determination of seawater diffusion constant as an example-10 aircraft. The diffusion constant is determined by directing a laser beam from an airplane to the surface of the sea and measuring its reflection.

Terrain can also be mapped using satellite imagery. Good large-scale sa-15 telltale images require multiple shots and interlocking. Current methods are still incomplete and the combination of multiple images is complicated. A large amount of information is obtained from images in electronic form. Extracting the correct and desired information from the information is difficult and requires a lot of complex computation. The best modeling: ·, · combines several different measurement methods. For example, data from satellite imagery can be combined with measurement results from y samples. However, reliable connection * 25 is a complicated operation.

PURPOSE OF THE INVENTION

It is an object of the invention to provide a new type of remote sensing method and to combine remote mapping results with other monitoring methods. In particular, it is an object of the invention to alleviate the above problems in satellite remote sensing. It is a further object of the invention to provide a method by which a large amount of '35 remote sensing material can be extracted with the desired information; formation.

M · | 3, 113894

SUMMARY OF THE INVENTION

The present invention relates to a method for remote sensing. The method of the invention generates a plurality of resolution-5 patterns in the tracking areas. A resolution model is a model that delimits the monitoring area to be addressed. The monitoring area refers to the geographical area to be measured. If the monitoring area is large or otherwise problematic, it may also consist of several smaller-10 resolution ranges. The resolution models for the monitoring areas are created using ready maps. In the case of watercourses, the resolution model can mean a boundary with a certain distance to the shore, for example 500 meters. The delimitation is not absolute, but pie-15 bands, islands and quarries can be included in the resolution range if they are so small that they do not significantly affect the measurement results. In the method of the invention, models are created at several different resolutions. For example, the models can have resolutions of 20, 1000, 500, 300 and 250 meters. The resolution models for the tracking regions are created based on the resolutions of the measurement instruments. In the resolution range model.- ··. one point always falls within one resolution range.

The resolution models created are combined with a resolution-25 region model. There are different types of resolution range model; / resolution ranges. The resolution range can also be segmented smaller to form tracking areas of desired size and shape. This is especially the case when measuring the sea. The tracking-30 areas are formed as a combination of individual sub-areas of the resolution range model. A single sub-area of the resolution range model may be part of several • * '· beach areas at the same time.

The created resolution range model is utilized in the analysis of remote 35 mapping images. Remote sensing images ··, for example, from satellite or airplane.

.... Remote sensing images are aligned with resolutions 113894 4 so that the satellite measurements of the areas to be tracked can easily be distinguished from the measured remote sensing material. The resolution area model can be used to remove areas near the shores of exemplary water bodies from remote sensing material. The actual monitoring is focused on the monitoring areas that are formed in the database by combining the appropriate monitoring areas in the desired area.

Once the resolution area model is aligned, the desired objects can be measured from the remote 10-size mapping image.

The remote sensing image can also be replaced with other measurement data if needed. Thanks to the resolution range model, the measurement is precisely targeted to the desired area according to the resolution of the measuring instrument. In the measurement, the digital measurement results of the satellite-15 connector observers are converted into physical radiation measurements. Preferably, the measurement results are spectral images in which each pixel is comprised of a spectrum reflected from a depicted point and measured at specific wavelengths of the measuring instrument. Spectral: · In addition to images, the system may also process measurement data. ·, · Other forms of measurement. The resulting spectra are sorted and classified. In this way, only the desired spectra are extracted from the spectra. The spectra to be erased include spectra, *. · Where disturbances occur. Examples of such interference are Cloudiness and Measuring Instrument Faults.

Classified and organized measurements can be analyzed immediately. Typically useful results are also archived for later processing. The archive also contains information on, for example, the measuring equipment used and the measurement situation. Round-*; . elaborate and organized measurement results are easy to archive. Archiving is well worth it because 35 analytical methods and analytical equipment evolve rapidly, and later the same measurement results 113894 5 can provide new insights into the benefits of better analysis.

The invention greatly improves remote sensing from satellite images. The method according to the invention can control a large number of images, so combining the observations does not become a problem. By the method of the invention, the measurement can be more precisely targeted to a desired area which reduces the size of the material to be processed. Further, classifying and arranging the spectra reduces the number of measurement results to be analyzed, thereby reducing the time needed for analysis.

LIST OF FIGURES

Fig. 1 illustrates a functional block diagram of an embodiment of the method of the invention, and Fig. 2 shows an example functional block diagram of an embodiment of a database system according to the invention, Fig. 3 illustrates an embodiment of an I system according to the invention; 1 resolution range model.

DETAILED DESCRIPTION OF THE INVENTION

Figure 1 shows a functional block diagram of a method according to the invention. For the method according to the invention, it is essential to create resolution models. Reso lution model refers to a model that determines at what resolution a measurement area can be collected from the observation area. When starting the analysis, check if the necessary templates and resolution range models exist, step 10. If no templates exist, they should be ____ created. Resolution models are created as vectors from the map by · · 35 defining, step 11. The resolution to be used · · · · · · · · · · · · · · · · · · · · · · · · · mainly depends on the measurement instrument used 6 113894. For example, the resolution can be 250 meters and 500 meters. In the case of watercourses, the resolution model may mean a boundary with 5 specific distances to the shore, for example 500 meters. The delimitation is not absolute, but small earth strips, islands and quarries can be included in the resolution range if they are so small that they do not significantly affect the measurement results. The determination of the distance is important 10 because the reflection typically caused by the terrain is significantly higher than the reflection of the water, so that even a small strip of terrain causes a large measurement error. Because areas that can be tracked at a resolution of 500 meters can also be tracked at a resolution of 250 meters, the tracking areas of instruments of different resolutions consist of combinations of different resolution ranges. Based on the resolution models of the created monitoring areas, a resolution area model is created, step 12. The created models are saved for later use 20. Different resolution, · resolution areas can be distinguished from the resolution range model. Figure 4 shows such a model.

In the resolution range model shown in Figure 4, a resolution of 250 meters is represented by horizontal lines and a resolution of 500 meters by vertical lines. The resolution area '25 can also be segmented smaller to form monitoring areas of desired size and shape. This is especially the case when measuring the sea. Finally, the desired monitoring areas are formed by combining them from resolution areas, step 13.

30 A single part of a resolution range model may fall into more than one monitoring area.

I., If there is a model for the object, the object can be imaged. "(Vata, step 14. For example, * ·· why images are used to represent the object; each pixel represents ·; · 35 spectra measured from the object. Satellite measurements can be processed either as a point database taking all measurements • · ···, or converting them into 113894 7 first by converting them to a traditional raster image that produces an image by selecting points in a given coordinate grid. Closest Measurements Typically, the area to be observed is large. one resolution range is sufficient to cover the entire observation area.

Once the subject is photographed, the resolution range model is superimposed on the subject to be analyzed, step 15. The model allows easy picking of measurements that hit areas of interest and thus potentially useful while the rest of the image is unusable, step 16. Measurement results are classified as good and bad, step 17. Poor measurement results include, for example, images in which Cloud 15 prevents reliable observations. Good measurement results are analyzed, step 18. If necessary, the measurement results and their interpretations are archived. Typically, useful measurement results found in individual resolution ranges are always archived, and completely invalid results are never archived.

Figure 2 shows an example of a database system for an \ application according to the invention. In the figure, blocks 20-22 illustrate satellite observation data, blocks 23 '··· - 26 illustrate metadata for measuring instruments, observations and results * - * 25, blocks 27-29 and 210-213 illustrate monitoring area data, blocks 214-217 illustrate block interpreting results, blocks 218-220 illustrate measurements and additional information, and block 221 the database functionality and data processing functions.

The satellite observation data 20-22 comprises the observation filtering information 20, the observation log 21, and the satellite observations 22 made by the measuring devices. The satellite observation 22 comprises basic measurement packages. The filtering data 20 may comprise a plurality of different filtering data; for example, by which algorithm and under what V conditions the information is filtered. This limits the amount of data packets that can be collected, since filtered I 113894 8 packets are not stored. Because of the high computational power required to process satellite observations, logs 21 are kept in areas where packets have already been processed. The method of processing packets 5 can also be stored in the log information.

Metadata 23 - 26 for Measuring Instruments, Observations and Results contains important information for the interpretation of measurement results. The metadata comprises satellite data preprocessing data 23 and observation type 24.

10 Satellite instrument data 25 is particularly useful when multiple satellite observations are combined into a single observation. The accuracy of the result also affects the interpretation. The accuracy of the result is expressed by the data type 26, which tells the scale of the point to be measured and the scale on which the output data describes it.

The monitoring area data 27-29 and 210-213 include all monitoring data relevant to the measurement. Resolution area information 27 comprises a set of minimum and maximum chromaticities of dynamics for each resolution area and a midpoint of the area. In addition, the reso- lution area information 27 comprises a unique identification of the area. ! the geographical location of the area. Multiple resolution ranges for larger compound ranges of resolution ranges *. * 25 using 28 merge data. Separate external identifiers 213 may also be used for the area, which may indicate, for example, that the area belongs to a municipal area. Tracking area data 210 is combined to form an entity. In addition to area measurement results 30, information 210 may be associated with communications with monitoring areas 211, such as belonging to the same body of water, or any other additional information. 212. Other possible additional information or connections are, for example, the neighborhood of the monitoring areas or the size of the flow: · 35. Where appropriate, satellite observations of observation areas may also be supplemented by locally-made observations. * Local observations may be, for example, water samples.

The interpretation results 216 are, of course, the most interesting part of the system mie-5 for the end user. The interpretation results of the observations may be accompanied by raster images 217, for example maps. Further, the interpretation results may comprise statistical indicators 214 from the data of the measured objects as well as a list 215 of the detected objects.

10 Observation times 218 are used to connect satellite observations with other information 219. Additional information 219 is, for example, measurements of the weather. Interpretation times 220 can be used to calculate averages of observations for a particular month, for example.

15 The various data are combined and processed by the database computing function 221. Such functions include, for example, the classification and the organizing functions. These self-defined functions allow the system administrator to provide information about ha-20 lesions to the system, or to map data to the system through the system, for example, by referring to ambiguous parts of the raster image where the monitoring area in question is displayed.

Figure 3 shows a system according to the invention. In the system of the invention, satellite 30 captures images of terrain 31. The terrain comprises bodies of water depicted by dotted line 311. Satellite 30 transmits observations made with dimensional instruments, such as spectral images, to satellite receiver 34. The Sa-30 satellite receiver 34 is connected to the information system. 34 are provided with all the necessary means for analyzing the spectra. Computer - *. *. the system comprises means 35 for generating resolution models of the monitoring regions and means 36 for creating a resolution model of 35: - *: * 35 regions from the resolution models of the monitoring regions. Resolution templates are created only once and then saved for later use. 113894 10 means for creating a model 36 comprises a digital map application to which the desired resolution is determined. The monitoring areas are created by the monitoring area definition means 312, which extract 5 resolution areas belonging to each monitoring area.

Described objects are targeted to the resolution range model by using the cursor 37. Different resolution ranges can easily be extracted from the focused image. The desired spectra are measured by measuring means 38 from different reso-10 lution areas. Since the resolutions of the measuring instruments are known, the measurement results outside the resolution range may be rejected by the classifier 311. The classifier may also classify the measurement results according to other rules 15. The measurement result may be discarded, for example, due to excessive cloudiness. Good measurement results are analyzed by analyzer 39. The system may, if necessary, incorporate field measurement results, which are collected by sampling means 33. Such samples 20 include, for example, water and soil samples. The aforementioned means 35-310 may be implemented as hardware, software or a combination thereof.

Figure 4 shows a resolution range model according to the invention. The model is placed on top of Figure 40.

25 The shoreline of the watercourse is represented by a curve 45. In the figure '· * the horizontal line represents a resolution of 250 meters and the vertical line of t · \ represents a resolution of 500 meters. In the figure, there are four 500-meter instrumented observations 41-44. The observation 41 is partially within the re-30 resolution range but is discarded because the center of gravity is outside the resolution range.

Typically, resolution ranges are defined as If the focal point is in the resolution range, you will not notice a disturbing amount of terrain. Observation 42 is not within the resolution range, so some of the measured • observation is terrain and measurement cannot be accepted. Observation 43 is completely out of range 11113894 and is discarded. Observation 44 is completely within the resolution range, so it is acceptable. Figure 4 also shows an alternative in which the resolution areas are divided into the tracking areas by a straight line 46. In this case, the areas above and below the line 46 can be viewed. Sat as their own monitoring areas or combine them as needed. Measurements are stored in the database according to the resolution ranges, so connection to the monitoring areas is only done at the analysis stage.

The invention is not limited solely to the above exemplary embodiments, but many modifications are possible within the scope of the inventive idea defined by the claims.

* 1 t t i I »» »{(| *« »» »t f I» »- · * 1 I · I» 1

Claims (21)

A method for analyzing remote mapping material, the method comprising the steps: a spectral image is captured by the object to be analyzed; a remote mapping image (40) is located; the remote mapping image (40) is analyzed; characterized in that the method further comprises the steps: 10 resolution models are created; of the resolution models, a resolution area model is created; the resolution area model is focused on the remote mapping material; From the remote mapping material, measurements are picked out to the desired area (41 - 44); and so the classified measurement results are analyzed. 2. A method according to claim 1, characterized in that the desired resolution areas are combined into a follow-up area. Method according to Claim 1 or 2, characterized in that the measurement results are classi »» I. | before the analysis. 4. A method according to claim 1, 2 or 3, characterized in that the resolution areas, *; are achieved from the map by drawing boundaries. , ·; Method according to any of the preceding claims 1-4, characterized in that the resolution area model is achieved by placing the resolution models of the follow-up areas on each other. , ···. 6. A method according to any one of the preceding claims 1-5, characterized in that by 'in the resolution area model, the distance map image is limited to the area to be analyzed according to the reso-,. The solution of a selected measuring device. Method according to any one of the preceding claims 1 - 6, characterized in that the remote mapping material is a spectral image. 8. A method according to any of the preceding claims 1 - 7, characterized in that the remote mapping material consists of a number of measurements. A method according to any of the preceding patents claims 1-8, characterized in that the measurement results are classified as approved and failed. 10. A method according to claim 9, characterized in that the measurement results are rejected if the object is outside the resolution range. Method according to any of the preceding claims 1 - 10, characterized in that the terrain measurement data is added to the remote mapping data. A system for analyzing remote mapping material, the system comprising: a satellite (30) for receiving a remote mapping image; ! a satellite receiver (34) for receiving said remote mapping image from the satellite; and: * a computer system (32) for analyzing the remote mapping image; 25, characterized in that the system | further includes:; means (35) for creating resolution models of the follow-up areas; means (36) for creating a resolution range model of the resolution models; an alignment device (37) for aligning the resolution area model to the remote mapping image; measuring equipment (38) for measuring desired spectra; and 3. an analyzer (39) for analyzing the • V classified measurement results. 17 1 13894 The system of claim 12, characterized in that the system further comprises means (312) for joining desired resolution regions to a follow-up region. j 5 System according to claim 12 or 13, characterized in that the system further comprises a classifier (310) for classifying the measurement results. System according to claims 12, 13 or 14, 10, characterized in that the computer system (32) is arranged to limit the resolution areas on the map. System according to any of the preceding claims 12 - 15, characterized in that the computer system (32) is arranged to combine the resolution models of the resolution [15] into a resolution area. desmodell. 17. A system according to claim 12 or 15, characterized in that the computer system (32) is arranged to limit the area to be adapted with the resolution area model from the remote mapping image; lysed. 18. A system according to any of the preceding claims 12-17, characterized in that the measuring equipment (38) is arranged to measure the desired spectra by taking a spectral image of the object. > f 19. A system according to any of the preceding patent claims: claims 14-18, characterized in that the classifier (310) is arranged to classify the measurement results as approved and failed. 20. A system according to claim 19, characterized in that the classifier (310) is * arranged to reject the measurement results if the object j is outside the resolution range. A system according to any of the preceding claims 12 - 20, characterized in that the computer system (32) is arranged to combine terrain measurement data with the remote mapping.