KR101703442B1 - Apparatus and method of extrcting a seamline for mosaicking of satellite images - Google Patents

Abstract

Provided is a method for generating a mosaic image by synthesizing a plurality of images by an image processing device. The method comprises: a step of extracting a normalized vegetation index of a reference image among the plurality of images; a step of performing a down sampling on the extracted normalized vegetation index; and a step of detecting the edges of the down-sampled normalized vegetation index to extract the shortest path connecting the edges as an initial seamline line for generating the mosaic image.

Description

[0001] APPARATUS AND METHOD OF EXTRACTION A SEAMLINE FOR MOSAICKING OF SATELLITE IMAGES [0002]

The present invention relates to a line extraction apparatus and method for satellite image mosaicking, and more particularly to a line extraction apparatus and method for reducing computation time and memory usage by using an initial line of intersection.

Satellite image mosaicking refers to an image processing technology that spatially synthesizes a plurality of images and generates a single image representing a wider area. The plurality of images will have different data values according to the photographing apparatus, the photographing geometry, the sunshine state, and the like. Therefore, the main purpose of the mosaicking technique is to minimize the difference between the two images to be mosaicked.

There are a number of studies on how to extract meaningful seamlines as baselines of mosaic kings. A joining line refers to a reference line when a plurality of different images are synthesized. It is pointed out that the conventional method of extracting the junction line that minimizes the difference between the pixel values of the two images passes through the main facility existing in the downtown area when applied to the high resolution satellite image. In addition, the slope method that extracts the path with the minimum slope of neighboring pixels as a bond line shows some good performance in the urban area, but when applied to a high resolution satellite image in which the number of pixels is increased, the calculation time is considerably delayed, There is a problem due to memory usage.

In addition, the conventional methods have a problem that the quality difference of the joining lines generated depending on the type of the region including the image, such as the natural area, the urban area, and the mixed area, is severe.

There is provided a method of extracting a Normal Distribution Vegetation Index (NDVI) and generating an initial seam line for mosaicking using the extracted normalized vegetation index. According to one aspect of the present invention, there is provided a method of generating a mosaic image by synthesizing a plurality of images by an image processing apparatus, the method comprising: extracting a normalized vegetation index of a reference image among the plurality of images; Sampling the edge of the downsampled normalized vegetation index and extracting a path connecting the edge through the shortest path as an initial line for generating the mosaic image.

According to an exemplary embodiment, performing the downsampling may include binarizing the normalized vegetation index using a predetermined threshold.

According to another embodiment, the step of extracting with the initial joint line may include inputting a pixel of an edge of the normalized vegetation index as one node, and calculating a line connecting the nodes through the shortest path using a dextral algorithm .

According to another embodiment, the method further comprises the steps of: detecting an edge of the reference image and generating a cost image that gives a smaller cost as the tilt magnitude is smaller; and generating the cost image using the cost image and the initial joint line, And extracting a final joining line to be generated. In addition, the step of generating the cost image may include generating a range cost image existing within a predetermined range based on the initial joint line. In addition, the step of extracting the final seam line may include extracting the shortest path as the final seam line that minimizes the cost by using the range coast image.

According to another aspect, there is provided a method of generating a mosaic image by combining a plurality of images by an image processing apparatus. The method includes extracting a normalized vegetation index of a reference image among the plurality of images, calculating a ratio of pixels of the reference image having the normalized vegetation index existing in an overlapping region of the plurality of images and equal to or less than a first threshold value And determining a method of extracting a joint line for generating the mosaic image according to the ratio of the pixels.

According to one embodiment, if the ratio of the pixels is less than a second threshold and less than a third threshold, the method further comprises: performing downsampling on the extracted normalized vegetation index, and comparing the downsampled normalized vegetation index And extracting a line connecting the edge of the normalized vegetation index with the shortest path as an initial line for generating the mosaic image. More specifically, if extraction of the initial junction line fails, the method may include determining whether the ratio of the pixels is less than or equal to a fourth threshold, and determining whether any of the normalized cross-correlation algorithm and slope comparison algorithm And a step of selecting the step.

According to another embodiment, if the ratio of the pixels is below a second threshold or exceeds a third threshold, the method further comprises: determining whether the ratio of the pixels is below a fourth threshold; , A normalized cross-correlation algorithm, and a slope comparison algorithm. In addition, the normalized cross-correlation algorithm may obtain a cross-correlation to the overlap region of the plurality of images, and calculate a path that minimizes the cross-correlation to a joint line. Meanwhile, the slope comparison algorithm may calculate, as a joint line, a path that minimizes the slope between neighboring pixels of the reference image among the plurality of images.

FIGS. 1A, 1B, 1C, and 1D are diagrams for explaining a process of extracting an initial joint line for satellite image mosaicking using the normalized vegetation index according to an exemplary embodiment.
2 is a flow chart of a method for extracting an initial bond line according to one embodiment.
FIGS. 3A, 3B and 3C are diagrams for explaining a process of generating a range cost image using an initial joint line. FIG.
4 is a flowchart of a method of extracting a final joining line using an initial joining line according to an embodiment.
5A and 5B are exemplary diagrams showing a final seam line and a mosaicking image generated according to an embodiment.
6 is a flowchart illustrating a final joint line extraction method according to an embodiment.

In the following, embodiments will be described in detail with reference to the accompanying drawings. However, the scope of the present invention is not limited or limited by these embodiments. Like reference symbols in the drawings denote like elements. The terms used in the following description are chosen to be generic and universal in the art to which they are related, but other terms may exist depending on the development and / or change in technology, customs, preferences of the technician, and the like. Accordingly, the terminology used in the following description should not be construed as limiting the technical thought, but should be understood in the exemplary language used to describe the embodiments.

Also, in certain cases, there may be a term chosen arbitrarily by the applicant, in which case the meaning shall be stated in the corresponding description. Therefore, the term used in the following description should be understood based on the meaning of the term, not the name of a simple term, and the contents throughout the specification.

FIGS. 1A, 1B, 1C, and 1D are diagrams for explaining a process of extracting an initial joint line for satellite image mosaicking using the normalized vegetation index according to an exemplary embodiment. Referring to FIG. 1A, a satellite image for performing a mosaic is shown. Illustratively, an image including various vegetation such as a green vegetation in the northeast direction is provided, and an image including a city area such as a road, a building, and the like may be provided in the southwest direction. However, the satellite image shown in FIG. 1A is only an exemplary description for facilitating understanding of the present invention, and does not limit or limit the scope of the present invention. In addition, the satellite image mosaicking method corresponding to this embodiment can be applied not only to the boundary between the urban area and the natural area, but also to the satellite image including only the urban area or the natural area.

Referring to FIG. 1B, an image obtained by extracting a Normal Distribution Vegetation Index (NDVI) using the satellite image shown in FIG. 1A is shown. The normalized vegetation index is a vegetation index for analyzing information such as distribution and activity of vegetation using remote satellite imagery. More specifically, the normalized vegetation index can be calculated using the following equation (1).

ρ _NIR represents the spectral reflectance of the satellite image in the near infrared region, and ρ _RED represents the spectral reflectance of the satellite image in the infrared region. More specifically, the normalized vegetation index can determine the ratio of vegetation existing in the satellite image using the spectral reflection characteristic inherent to the plant. Illustratively, chlorophyll pigments present in plant leaves strongly absorb wavelength bands of 0.45 μm to 0.67 μm, and strongly reflect wavelength bands of 0.74 μm to 1.3 μm. Using the characteristics of the above plants, it is possible to detect the presence of vegetation with high absorption in the infrared region and strong reflectivity in the near-infrared region, and thereby analyze the characteristics of the region. NDVI is a value that is less than or equal to 1.0 and has a range that is greater than or equal to -1.0.

Accordingly, in this embodiment, an initial joint line using the above normalized vegetation index is extracted, and a final joint line for generating a mosaicking satellite image is calculated. Accordingly, according to this embodiment, the boundary lines of the natural environment and the urban environment can be set as the initial fixing lines, and natural topography such as mountains and rivers or artificial terrains such as dams, bridges and buildings can be separated by the mosaic Can be prevented.

Referring to FIG. 1C, an image obtained by down sampling the normalized vegetation index generated as shown in FIG. 1B is provided. As described above, each pixel of the satellite image may have a normalized vegetation index ranging from -1.0 to less than 1.0. In addition, the image processing apparatus according to the present invention can downsample the normalized vegetation index to further increase the processing speed. More specifically, the image processing apparatus can binarize the normalized vegetation index based on a predetermined threshold value. Illustratively, when the image processing apparatus sets a threshold value of a predetermined normalized vegetation index to 0.0, pixels with a normalized vegetation index of 0.0 or more will be sampled as 1, and pixels with a normalized vegetation index of 0.0 or less will be sampled as 0.

Downsampling performed to increase the processing speed and efficiency of voice or image files generally tends to involve some degree of trade-off in quality degradation. Considering the conventional joint line generation methods, raster-based data processing extracts joint lines by considering all pixel ranges, so that even though quality is not deteriorated, high-resolution satellite images, Time and exponentially increasing memory capacity, which is a problem in that it can not be practically applied.

However, according to the present embodiment, the normalized vegetation index is data used for extracting an initial joint line, and even if downsampling such as binarization is performed, it is possible not to significantly affect the quality of the final joint line ultimately derived. Accordingly, it is possible to increase the processing speed by generating an initial joint line by downsampling, and to reduce the memory usage for final joint line creation by setting a range of pixels to be processed for final joint line generation around the initial joint line have.

Referring to FIG. 1D, an exemplary illustration of an initial seam line extracted using a normalized vegetation index of a downsampled satellite image is shown. Comparing the shape of the bonded initial bond line with the previously described FIG. 1A, it can be seen that the boundary line between the natural area and the urban area is extracted as the initial bond line. According to this embodiment, it is possible to guarantee a high possibility that the main terrain of the natural area or the main terrain of the urban area of the city is not separated by the seam line. A detailed description of the construction of the algorithm for extracting the initial joint line will be described together with the drawings to be added below.

2 is a flow chart of a method for extracting an initial bond line according to one embodiment. A method 200 for extracting an initial line of intersection includes extracting 210 a normalized vegetation index of a plurality of images, performing down-sampling on the extracted normalized vegetation index 220, And extracting an initial joint line 230 using an edge of the vegetation index.

Step 210 is a step of extracting a normalized vegetation index of a reference image from a plurality of images. In step 210, a reference image from which an initial joint line is extracted may be selected. In addition, in step 210, an area in which the reference image and the remaining plurality of images overlap can be determined. In addition, the normalized vegetation index of the overlapping area may be extracted in step 210. More specifically, the normalized vegetation index can be calculated using Equation 1 described above.

Step 220 is a step of performing downsampling on the extracted normalized vegetation index. More specifically, in step 220, the normalized vegetation index may be binarized based on a predetermined threshold. In one embodiment, the predetermined threshold may be determined as a normalized vegetation index of 0.0. Experimentally, if the water status of the vegetation is not considered as a variable, the area where the vegetation is present has a normalized vegetation index of 0.1 to 0.6. Accordingly, when the predetermined threshold value is set to 0.0, the pixel corresponding to the urban area has a normalized vegetation index of 0, and the pixel corresponding to the natural area can have 1 as the normalized vegetation index.

In the case of performing down-sampling on the satellite image, a loss of the necessary multiple information will be expected. Therefore, in the utilization of normalized vegetation index, downsampling such as binarization will not be able to analyze data to a meaningful extent in widely used fields such as photosynthesis degree of plant, soil condition analysis, and drought prediction. However, according to the embodiment of the present invention, the normalized vegetation index can be used to extract an initial joint line that generates a range for extracting a final joint line in satellite image mosaicking. More specifically, the initial bond line may be present at the boundary of the area where natural and urban areas are mixed, in order to prevent natural terrain or artificial terrain from being separated by the bond line. Considering the above points, according to the embodiment of the present invention, since a significant initial joint line is obtained as result data while reducing the amount of operation by downsampling on the normalized vegetation index, a remarkable effect can be expected .

Step 230 extracts an initial line of intersection using the downsampled normalized vegetation index. More specifically, the path connecting the edge of the normalized vegetation index to the shortest path can be extracted as an initial joint line for generating a mosaic image. More specifically, in step 230 a Dijkstra Algorithm may be used to extract the initial joint line.

The extreme algorithm is an algorithm that finds the shortest path from the starting node to reach the destination node. However, in the present embodiment, a multi-extrude algorithm can be applied by treating pixels at neighboring pixels and edges between pixels as one node. Illustratively, the first pixel in the natural region will have a normalized vegetation index of one, and the second pixel in the urban area will have a normalized vegetation index of zero. Accordingly, an edge can be generated in a region where the first pixel and the second pixel are adjacent to each other. The pixels of the edges generated as above can be considered as one node, and the shortest path from one node to the destination node across the satellite image can be calculated as an initial joint line.

The description of steps 210, 220, and 230 described above may be applied to an image processing apparatus including a processor. The image processing apparatus can extract an initial joint line using the normalized vegetation index using the above description.

FIGS. 3A, 3B and 3C are diagrams for explaining a process of generating a range cost image using an initial joint line. FIG. FIG. 3A shows a cost image generated using the gradient technique. The image processing apparatus according to the present embodiment can compare pixel values of overlapping regions of a plurality of images to be hatched. The image processing apparatus can calculate a slope according to pixel values of a plurality of images. In addition, the image processing apparatus can calculate the slope, size and direction of the discontinuous portion of the pixels, and edge localize the portion where the slope becomes maximum locally. The image processing apparatus can generate a cost image by giving a small cost to the detected edge portion and giving a high cost to a portion in which a difference between a plurality of pixel values is small.

Referring to FIG. 3B, a buffer zone centered on the initial bond line and the initial bond line calculated in step 230 is shown. The buffer area indicates a range for the image processing apparatus to compare pixel values of a plurality of images to calculate a final joint line. More specifically, the image processing apparatus can set a buffer area using statistical data such as average, variance, and mode values of a plurality of images.

3C shows the range cost image generated by extracting the cost image as much as the buffer area. The image processing apparatus can extract a final joining line for mosaicking a plurality of images by applying a joining line extraction method to the range cost image. In one embodiment, the joint line extraction method is a minimum absolute gray difference sum (MAGDS) algorithm that consecutively finds a point where the absolute value of the difference is minimum after summing the neighboring pixel values at the same position in two overlapping images . In another embodiment, the joint line extraction method may be an NCC algorithm that compares a normalized cross correlation (NCC) of a plurality of images and finds a path that minimizes the cross correlation coefficient. As another embodiment, the above-described multi-extensional algorithm can be applied to the method of extracting the line of intersection. The image processing apparatus can use the pixels in the range cost image to determine the pixels of the edge as one node and calculate the shortest path from one node to the destination node across the satellite image as the final joint line.

4 is a flowchart of a method of extracting a final joining line using an initial joining line according to an embodiment. A method 400 for extracting a final seam line using an initial seam line includes generating a cost image by detecting an edge of a reference image 410 and generating a mosaic image using the generated cost image and the initially extracted seam line (Step 420). ≪ / RTI >

In step 410, the image processing apparatus can detect the edge of the reference image. In one embodiment, the image processing apparatus extracts statistical data such as average, variance, and mode values of a plurality of satellite images to be mosaicked, assigns weights according to the importance of the statistical data, And the reference image can be selected. In addition, in step 410, the image processing apparatus can generate a cost image that gives a smaller cost as the magnitude of the slope of the pixel of the reference image, using the detected edge.

In operation 420, the image processing apparatus may extract a final joint line for generating the mosaic image using the generated cost image and the initial joint line. In addition, step 420 may include generating a range cost image that is within a predetermined range based on the initial bond line. In step 420, the image processing apparatus selects pixels included in the range cost image in the overlap area of the plurality of images, detects the shortest path that minimizes the cost in accordance with the range cost image, And can be extracted as a final joining line.

5A and 5B are exemplary diagrams showing a final seam line and a mosaicking image generated according to an embodiment. Referring to FIG. 5A, a first mosaicking image 510 and a first final joining line 520 generated using the reference image shown in FIG. 1A and the like are shown. As described above, the reference image shows a mixed area including both a natural area having a high normalized vegetation index and a city area having a low normalized vegetation index. Looking at the first final seam line 520, it can be seen that the natural and urban areas are extracted in the form of a boundary line that traverses from the south-west direction to the northeast direction without interrupting both the natural area and the urban area. Accordingly, a joining line 520 that does not cut a ridge, a mountainous terrain, or a main building in the city center included in the reference image is generated, and a mosaicking satellite image based thereon can be obtained. In addition, compared to the slope technique in which the slopes of all the pixels in the conventional overlap area were compared, embodiments of the present invention can utilize a range cost image based on the initial bond line to extract the final bond line You can do it.

Referring to FIG. 5B, a second final joining line 540 generated in a natural area having a high normalized vegetation index and a second mosaicking image 530 generated using the second final joining line 540 are shown. Referring to the second final joining line 540, it can be confirmed that the extracted information is extracted along with valid information like a valley formed in a mountainous area. In contrast to the conventional method, it is possible to confirm the effect that the final bonding line 540 is generated while preserving the formed information without irregularly locating the center of the acid.

6 is a flowchart illustrating a final joint line extraction method according to an embodiment. The final joint line extraction method 600 extracts the histogram of the normalized vegetation index of the reference image and analyzes the extracted histogram to differentiate the joint line extraction according to the environment included in the reference image. Accordingly, by using the initial joint line extraction method using the normalized vegetation index according to the present embodiment, a normalized cross-correlation algorithm is used in a natural area rich in vegetation, Finally, we show a method of extracting a hybrid bond line using an optimal technique according to the situation using a slope technique in the case of a dry soil or an urban area having little vegetation.

Step 610 extracts the histogram of the normalized vegetation index of the reference image among the plurality of images. In step 610, the description of step 210 described above may be applied. In addition, in step 620, the image processing apparatus can determine whether the ratio of the pixel corresponding to the predetermined condition is equal to or greater than the second threshold value and less than the third threshold value. More specifically, the image processing apparatus can set the normalized vegetation index of the pixel to be equal to or less than the first threshold value as the predetermined condition. In one embodiment, the first threshold may be set to 0.0. In general, if the normalized vegetation index is less than or equal to 0.0, it is recognized as an urban area. In step 620, it is determined whether the reference image is a natural region, an urban area, or an area using the histogram of the normalized vegetation index of the reference image. It is a step of judging whether or not it is a mixed area.

(630) performing downsampling on the extracted normalized vegetation index if the ratio of the pixels corresponding to the predetermined condition is determined to be equal to or less than the second threshold value and equal to or less than the third threshold value in step 620, A step 640 of extracting an initial joint line using the edge of the vegetation index may be performed. That is, if it is determined in step 620 that the reference image corresponds to the mixed region, the method 200 for extracting the initial joint line described above may be applied.

In addition, step 650 is a step of determining whether or not the extraction of the initial joint line has been successful through steps 630 and 640. If the edge of the pixel along the normalized vegetation index has not been generated sufficiently, the image processing apparatus may fail to extract the initial joint line despite steps 630 and 640. However, if an initial joint line is generated in step 650, step 660 may be performed. Step 660 is a step of extracting a final seam line, which includes generating 410 a cost image by detecting an edge of the reference image described above with reference to FIG. 4, and generating the mosaic image using the cost image and the initial seam line (Step 420). ≪ / RTI >

On the other hand, if it is determined in step 620 that the ratio of the pixel corresponding to the predetermined condition is less than the second threshold value or exceeds the third threshold value, or if it is determined in step 650 that the extraction of the initial joint line has failed, 670) may be performed. Step 670 is a step of determining whether the ratio of the pixel corresponding to the preset condition is equal to or less than a fourth threshold value. That is, in step 670, the image processing apparatus determines whether the reference image is close to the natural area or close to the urban area.

According to the result of the determination, if the pixel corresponding to the predetermined condition is equal to or less than the fourth threshold value, the normalized cross-correlation algorithm 680 may be applied to extract the final joint line. More specifically, the normalized cross-correlation algorithm 680 calculates an intersection correlation value of pixels included in two different images, and shows an algorithm for extracting a shortest path as a final intersection line so that the cross-correlation is minimized.

On the other hand, if the pixel corresponding to the predetermined condition is determined to exceed the fourth threshold value, the slope comparison algorithm 690 may be applied to extract the final fixing line. More specifically, the slope comparison algorithm 690 compares slopes of pixel values of two different images, detects a pixel having a locally large slope as an edge, and connects the shortest path to prevent the edge from being detected Respectively. In general, artificial terrain, such as buildings or bridges, will have a large change in slope value at that boundary and will be detected as an edge. Thus, according to the slope comparison algorithm, the shortest path will be searched to avoid the boundaries of the building or bridge, resulting in a final bond line that does not cross the main terrain.

The embodiments described above may be implemented in hardware components, software components, and / or a combination of hardware components and software components. For example, the devices, methods, and components described in the embodiments may be implemented within a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array such as an array, a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For ease of understanding, the processing apparatus may be described as being used singly, but those skilled in the art will recognize that the processing apparatus may have a plurality of processing elements and / As shown in FIG. For example, the processing apparatus may comprise a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of the foregoing, and may be configured to configure the processing device to operate as desired or to process it collectively or collectively Device can be commanded. The software and / or data may be in the form of any type of machine, component, physical device, virtual equipment, computer storage media, or device , Or may be permanently or temporarily embodied in a transmitted signal wave. The software may be distributed over a networked computer system and stored or executed in a distributed manner. The software and data may be stored on one or more computer readable recording media.

The method according to an embodiment may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced. Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (14)

A method of generating a mosaic image by combining a plurality of images by an image processing apparatus, the method comprising:
Extracting a normalized vegetation index of a reference image among the plurality of images;
Performing downsampling on the extracted normalized vegetation index;
Sampling an edge of the downsampled normalized vegetation index and extracting a path connecting the edge through the shortest path as an initial line of intersection;
Setting a buffer area for comparing pixel values of the plurality of images using the initial splitting line;
Detecting an edge of the reference image and generating a cost image giving a smaller cost as the tilt size is smaller, corresponding to the size of the buffer area; And
Extracting a final seam line for generating the mosaic image using the cost image, the initial seam line, and the plurality of images;
≪ / RTI > The method according to claim 1,
Wherein performing the downsampling includes binarizing the normalized vegetation index using a predetermined threshold. The method according to claim 1,
Wherein the step of extracting with the initial joint line includes inputting pixels of an edge of the normalized vegetation index as one node and calculating a line connecting the nodes through a shortest path using a dextral algorithm. delete The method according to claim 1,
Wherein generating the cost image comprises generating a range cost image that is within a predetermined range based on the initial bond line. 6. The method of claim 5,
Wherein the step of extracting the final joining line comprises extracting the shortest path to the final joining line using the range cost image to minimize the cost. A method of generating a mosaic image by combining a plurality of images by an image processing apparatus, the method comprising:
Extracting a normalized vegetation index of a reference image among the plurality of images;
Calculating a ratio of pixels of the reference image existing in the overlapping region of the plurality of images and having the normalized vegetation index lower than a first threshold value; And
Determining a method of extracting a joining line for generating the mosaic image according to the ratio of the pixels
Lt; / RTI >
And when the ratio of the pixels is less than the second threshold value and less than the third threshold value,
The method
Performing downsampling on the extracted normalized vegetation index; And
Sampling an edge of the downsampled normalized vegetation index and extracting a line connecting an edge of the normalized vegetation index as a shortest path as an initial line for generating the mosaic image
≪ / RTI > delete 8. The method of claim 7,
If the extraction of the initial seam line fails,
Determining whether the ratio of the pixels is less than a fourth threshold; And
Selecting one of the normalized cross-correlation algorithm and the slope comparison algorithm according to a result of the determination;
≪ / RTI > 8. The method of claim 7,
If the ratio of pixels is below a second threshold or above a third threshold,
Determining whether the ratio of the pixels is less than a fourth threshold; And
Selecting one of the normalized cross-correlation algorithm and the slope comparison algorithm according to a result of the determination;
≪ / RTI > 11. The method of claim 10,
Wherein the normalized cross-correlation algorithm obtains a cross-correlation in the overlap region of the plurality of images, and calculates a path that minimizes the cross-correlation as a joint line. 11. The method of claim 10,
Wherein the slope comparison algorithm calculates a path that minimizes the slope between neighboring pixels of the reference image among the plurality of images as a joint line. A computer-readable recording medium containing a program for executing a method for generating a mosaic image by synthesizing a plurality of images, the program comprising:
A command set for extracting a normalized vegetation index of a reference image among the plurality of images;
A set of instructions for performing downsampling on the extracted normalized vegetation index;
An instruction set for detecting an edge of the downsampled normalized vegetation index and extracting a line connecting an edge of the normalized vegetation index as a shortest path as an initial fixed line;
A command set for setting a buffer area for comparing pixel values of the plurality of images using the initial fixing line;
An instruction set for detecting an edge of the reference image and generating a cost image giving a smaller cost as the tilt size is smaller corresponding to the size of the buffer area; And
Extracting a final seam line for generating the mosaic image using the cost image, the initial seam line, and the plurality of images,
Readable recording medium. A computer-readable recording medium containing a program for executing a method for generating a mosaic image by synthesizing a plurality of images, the program comprising:
A command set for extracting a normalized vegetation index of a reference image among the plurality of images;
A set of instructions for calculating a ratio of pixels of the reference image existing in an overlapping region of the plurality of images and having the normalized vegetation index below a first threshold; And
Determining a method of extracting a joining line for generating the mosaic image according to the ratio of the pixels,
Lt; / RTI >
If the ratio of the pixels is greater than or equal to the second threshold and less than the third threshold,
A set of instructions for performing downsampling on the extracted normalized vegetation index; And
Sampled normalized vegetation index and extracting a line connecting an edge of the normalized vegetation index with a shortest path as an initial line of intersection.

Images