KR101823077B1 - Apparatus and method for obtaining high resolution image - Google Patents

Abstract

A high-resolution image acquiring apparatus acquires a plurality of frame images by photographing a moving object, estimates the relative motion of a current image with respect to a previous image as a coordinate value in decimal pixel units using the plurality of frame images, Resolution image is selected based on the coordinate values of the selected image and a high-resolution image is generated from the selected image using the super-resolution technique.

Description

[0001] APPARATUS AND METHOD FOR OBTAINING HIGH RESOLUTION IMAGE [0002]

The present invention relates to an apparatus and method for acquiring high resolution images, and more particularly, to a high resolution image acquisition apparatus and method for generating a high resolution image using a plurality of low resolution images acquired by photographing a moving object.

If the ladle of a steel manufacturing factory is used for a long time, the refractory brick which protects the ladle frame from hot molten steel may be deteriorated, and the ladle may be accidentally opened. The deterioration of the ladle is usually first generated locally, which can be known in advance by sensing the temperature rise outside the ladle.

The external appearance of the ladle can be photographed with an infrared camera to detect the area where the ladle is deteriorated. In order to accurately detect the degradation of ladders, a high-resolution image is required.

Recently, the utilization of the imaging system is increasing and the resolution of the image is increasing. However, in the thermal image field, there are various restrictions on the acquisition of the high resolution image, and it is expensive to increase the resolution differently from the CCD image.

The object of the present invention is to provide a high-resolution image acquiring apparatus and method capable of generating a high-resolution image using an image of a moving object photographed by a thermal imaging camera.

According to one embodiment of the present invention, a high-resolution image acquisition method of a high-resolution image acquisition apparatus is provided. A method for acquiring a high resolution image comprises the steps of acquiring a plurality of frame images by photographing a moving object, estimating a relative motion of a current image with respect to a previous image as a coordinate value in decimal pixel units using the plurality of frame images, Selecting an image to which a super resolution technique is to be applied based on coordinate values of the respective images, and generating one high resolution image from the selected image using the super resolution technique.

Wherein the step of selecting includes the steps of generating a unit pixel having a coordinate value of each of the images as a center point based on the number of images corresponding to coordinate values displayed in a plurality of unit pixels Dividing the selected unit pixel into a plurality of blocks, and selecting one image in the block in which the image exists among the plurality of blocks as an image to which the super resolution technique is to be applied . ≪ / RTI >

The step of selecting one image in the block in which the image exists may include selecting an image having a coordinate value closest to the center of each block in which the image exists, as an image to which the super resolution technique is to be applied .

The selecting of the unit pixel may include selecting a unit pixel having the largest number of images among the plurality of unit pixels.

The unit pixel may have a size that is set vertically, horizontally, and vertically with a coordinate value of the corresponding image as a center point.

Wherein the estimating includes estimating a relative motion of the current image as a coordinate value in decimal pixel units using an ECC (Enhanced Correlation Coefficient) based algorithm.

The plurality of frame images may include an image obtained by a thermal imaging camera.

The object may comprise a ladle used in a steel process.

According to another embodiment of the present invention, there is provided a high-resolution image acquisition apparatus for generating a high-resolution image using a plurality of low-resolution images. The high-resolution image acquisition apparatus includes an image capturing unit, a motion estimating unit, an image selecting unit, and a high-resolution image generating unit. The image capturing unit captures a moving object and acquires a plurality of frame images. The motion estimator estimates a coordinate value of the decimal point pixel of the relative motion of the current image with respect to the previous image using the plurality of frame images. The image selection unit selects an image to which the super resolution technique is applied based on the coordinate values of the estimated images. The high-resolution image generation unit generates a high-resolution image from the selected image using the super resolution technique.

Wherein the image selecting unit selects one unit pixel based on the number of images corresponding to the respective coordinate values displayed in the plurality of unit pixels having the coordinate values of the respective images as the center point, The image to which the super resolution technique is applied can be selected.

The image selecting unit may divide the selected unit pixel into a plurality of blocks and select an image having a coordinate value closest to the center of each block in which the image exists among the plurality of blocks as an image to which the super resolution technique is applied.

The image selection unit may select a unit pixel having the largest number of images among the plurality of unit pixels.

The image capturing unit may include an infrared camera.

The object may include ladders used in steel processing.

According to the embodiments of the present invention, a high-resolution image can be obtained using a single thermal imaging camera, and the refractoriness of ladders can be accurately detected using a high-resolution image.

1 is a block diagram illustrating a high-resolution image acquisition apparatus according to an embodiment of the present invention.
2 is a flowchart illustrating a method for acquiring a high-resolution image according to an embodiment of the present invention.
FIG. 3 is a flowchart illustrating an image selection method for applying a super resolution technique according to an embodiment of the present invention.
4 is a diagram illustrating motion information of a continuous frame image estimated by the motion estimation unit shown in FIG.
FIG. 5 is an enlarged view of the unit pixel A shown in FIG.
6 is a schematic diagram of a high resolution image acquisition apparatus according to another embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and similar parts are denoted by like reference characters throughout the specification.

Throughout the specification and claims, when a section is referred to as "including " an element, it is understood that it does not exclude other elements, but may include other elements, unless specifically stated otherwise. Also, the terms "part," "unit," "module," "block," and the like in the specification mean units for processing at least one function or operation, Can be implemented.

Now, an apparatus and method for acquiring a high-resolution image according to an embodiment of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram of a high-resolution image acquisition apparatus according to an embodiment of the present invention, and FIG. 2 is a flowchart illustrating a high-resolution image acquisition method according to an embodiment of the present invention.

1, the high resolution image acquisition apparatus 100 includes an image capturing unit 110, a motion estimating unit 120, an image selecting unit 130, and a high resolution image generating unit 140. The high resolution image acquisition apparatus 100 may further include an image storage unit 150.

Referring to FIG. 2, the image capturing unit 110 captures an object to obtain a continuous frame image (S210). The image capturing unit 110 may include an infrared camera. The frame image obtained by the thermal imaging camera may be a low resolution image. In an embodiment of the present invention, the object may be a ladle used in a steel process. The ladle contains molten steel. The inner side of the ladle is enclosed by a wall structure, ie a refractory brick, to protect the ladle, and the ladle containing molten steel moves to a place for processing.

The image capturing unit 110 captures the moving ladle and transmits the captured image to the motion estimator 120. Hereinafter, it is assumed that the object is a ladle.

The motion estimation unit 120 estimates the motion of the ladder from the continuous frame image acquired by the imaging unit 110 (S220). In addition to movement to the process site, there are fine movements due to vibrations from the plant. Therefore, the motion estimation unit 120 can estimate the motion of the ladder using the ECC (Enhanced Correlation Coefficient) -based algorithm from the images acquired by the image capturing unit 110 to the coordinate value of the decimal pixel unit. The ECC-based algorithm is an algorithm that estimates the motion of decimal pixels between images.

The motion estimation unit 120 compares the current image with an ideal image by an ECC-based algorithm, and estimates the relative motion of the current image with respect to the previous image as a coordinate value of the decimal pixel unit. The motion estimator 120 sets a coordinate value of a first image in a basic coordinate of a pixel unit consisting of an X axis and a Y axis and calculates a relative motion of a second image based on a coordinate value of a first image with respect to a second image, Lt; RTI ID = 0.0 > pixel value < / RTI > The motion estimation unit 120 displays the decimal pixel value of the relative motion of the current image on the basis of the coordinate value of the previous image in the basic coordinates of the pixel unit.

The image selection unit 130 selects an image to which a super resolution technique is to be applied using the motion of the ladle estimated by the motion estimation unit 120 at step S230. Specifically, the video selecting unit 130 generates a unit pixel having a predetermined size centered on each image based on all the images displayed in the basic coordinates in units of pixels by the motion estimating unit 120. Based on the coordinate positions of the displayed image in the unit pixel, the image selection unit 130 selects the unit pixels having the largest number of coordinate positions and selects an image to which the super resolution technique is applied in the unit pixel.

The high-resolution image generation unit 140 applies a super resolution technique to the images selected by the image selection unit 130 to generate a single high-resolution image (S240). Since a super resolution technique for generating a single high resolution image using a plurality of low resolution images corresponds to a known technique, a detailed description thereof will be omitted.

The image storage unit 150 may store an image selected by the image selection unit 130. [

FIG. 3 is a flowchart illustrating an image selection method for applying a super resolution technique according to an exemplary embodiment of the present invention. FIG. 4 is a flowchart illustrating motion information of a continuous frame image estimated by the motion estimation unit shown in FIG. FIG. FIG. 5 is an enlarged view of the unit pixel A shown in FIG.

Referring to FIG. 3, the image selection unit 130 sets a predetermined size around the coordinate value of each image for all the images in which the motion information of each image is represented by the coordinate value of the decimal pixel unit in the basic coordinates of the pixel unit (S310).

Motion information of all images estimated by the motion estimation unit 120 using the ECC-based algorithm can be displayed as coordinate values in decimal pixel units as shown in FIG.

Referring to FIG. 4, the coordinate value of the first image is set to (0, 0). From the second image to the last image, the decimal pixel value for the relative motion of the current image can be displayed based on the coordinate value of the previous image.

FIG. 4 is a decimal point pixel value for movement of each image estimated based on an ECC-based algorithm, and a decimal point pixel value for movement of each image is connected by a line.

4, it can be estimated that the ladle moves down to the left when it reaches the process site. Also, the movement of the ladle is shifted by about 55 pixels in the left direction and 4.5 pixels in the downward direction when analyzing based on the last image based on the coordinate value of the first image.

The image selection unit 130 generates a unit pixel of a predetermined size based on the coordinate values of the respective images displayed as shown in FIG. 4. For example, the image selection unit 130 generates unit pixels of 0.5 Unit pixel having a size of 1 * 1 can be generated. When the number of frame images obtained by the image capturing unit 110 is 500, motion information of 500 frame images is displayed as coordinate values as shown in FIG. 4, 500 units of coordinates Pixels can be generated.

The image selecting unit 130 checks the number of images with respect to the coordinate values displayed in the unit pixels generated with the center values of the coordinate values of the respective images.

The image selecting unit 130 selects a unit pixel having the largest number of images among a plurality of unit pixels based on the number of images with respect to the coordinate values displayed in the unit pixel (S320). The super resolution technique uses information of subpixels, ie, images with motion in decimal units. The pixel motion information in the integer unit is information that is overlapped with each other, and therefore does not contribute to the improvement of picture quality. Therefore, in order to select motion images of subpixel units, it is advantageous to apply the super resolution technique to select many points having images having motion of one pixel or less. Accordingly, the image selecting unit 130 selects a unit pixel having the largest number of images among a plurality of unit pixels.

The fact that the number of images in the unit pixel is large means that the movement of the ladle is minimized and the movement of the ladle is minimized. There is almost no difference between the pixel values between the frames. Therefore, a step of removing overlapping frame images is required.

The image selecting unit 130 divides the selected unit pixel into a plurality of blocks (S330).

In order to minimize overlapping frame images, the image selecting unit 130 selects images at one coordinate position to which the super resolution technique is to be applied, in a block in which there are images among a plurality of blocks, at step S340. The image selecting unit 130 can select an image at a coordinate position closest to the center of each block as an image to which the super resolution technique is applied.

The unit pixel A selected based on the decimal pixel values for the motion of each image shown in FIG. 4 is shown in FIG. In Fig. 5, "points" represent respective images indicated by decimal pixel values.

5, the image selecting unit 130 divides the unit pixel A into 16 blocks, and selects images of one coordinate position from each of the 16 blocks. (A1, a2, a3, a4, a5, a6, and a7) at the coordinates closest to the center of each block in the block to which the super resolution technique is applied.

The image selecting unit 130 stores the images to which the super resolution technique is applied in the image storing unit 150 (S350).

By using this image selection method, it is possible to effectively select the image to be applied to the super resolution technique while eliminating the overlap image. By applying the selected image to the super resolution technique, the processing speed can be improved, Can be obtained.

At least some functions of the high resolution image acquisition apparatus and method according to the embodiments of the present invention described above can be implemented in hardware or software combined with hardware. Hereinafter, an embodiment in which a high-resolution image acquisition apparatus and method are combined into a computer system will be described in detail with reference to FIG.

6 is a schematic view of a high resolution image acquisition apparatus according to another embodiment of the present invention. The motion estimation unit 120, the image selection unit 130, the high resolution image generation unit 140 And an image storage unit 150. The image storage unit 150 may be used to perform at least some of the functions of the image storage unit 150 and the image storage unit 150. [

6, a high resolution image acquisition device 600 includes a processor 610, a memory 620, at least a storage device 630, an input / output (I / O) interface 640, and a network interface 650 ).

The processor 610 may be implemented as a central processing unit (CPU) or other chipset, microprocessor, etc., and the memory 620 may be a dynamic random access memory (DRAM), a Rambus DRAM DRAM, RDRAM), synchronous DRAM (synchronous DRAM, SDRAM), static RAM (SRAM), and the like. The storage device 630 may be a hard disk, a compact disc read only memory (CD-ROM), a compact disc rewritable (CD-RW), a digital video disc ROM (DVD-ROM) , An optical disk such as a blu-ray disk, a flash memory, various types of RAM, or a permanent or volatile storage device. I / O interface 640 also allows processor 610 and / or memory 620 to access storage device 630 and network interface 650 may be coupled to processor 610 and / To access the network.

In this case, the processor 610 loads a program command for implementing at least some functions of the functions of the motion estimation unit 120, the image selection unit 130 and the high-resolution image generation unit 140 into the memory 620, The function of the image storage unit 150 may be placed in the storage device 630 so that the operations described with reference to FIGS. The program command may be stored in the storage device 630 or may be stored in another system connected to the network.

The processor 610, the memory 620, the storage device 630, the I / O interface 640, and the network interface 650 shown in FIG. 6 may be implemented in one computer or distributed in a plurality of computers .

The embodiments of the present invention are not limited to the above-described apparatuses and / or methods, but may be implemented through a program for realizing functions corresponding to the configuration of the embodiment of the present invention or a recording medium on which the program is recorded, Such an embodiment can be readily implemented by those skilled in the art from the description of the embodiments described above.

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, It belongs to the scope of right.

Claims (14)

A high resolution image acquisition method of a high resolution image acquisition apparatus,
Capturing a moving object to obtain a plurality of frame images,
Estimating a relative motion of a current image with respect to a previous image as a coordinate value of a decimal pixel unit using the plurality of frame images,
Selecting an image to which a super resolution technique is to be applied based on the coordinate values of the estimated images, and
Generating one high-resolution image from the selected image using the super resolution technique
/ RTI >
The step of selecting
Selecting one unit pixel based on the number of images corresponding to the coordinate values displayed in the plurality of unit pixels generated with the center point of the coordinate value of each image,
Dividing the selected unit pixel into a plurality of blocks, and
And selecting one of the plurality of blocks as an image to which the super resolution technique is to be applied. delete The method of claim 1,
Wherein the step of selecting one image in the block in which the image exists includes a step of selecting an image having a coordinate value closest to the center of each block in which the image exists, as an image to which the super resolution technique is to be applied, Acquisition method. The method of claim 1,
Wherein the selecting of the unit pixels includes selecting a unit pixel having the largest number of images among the plurality of unit pixels. The method of claim 1,
Wherein the unit pixel has a size set up, down, left, and right with a coordinate value of the image as a center point. The method of claim 1,
Wherein the estimating includes estimating a relative motion of the current image as a coordinate value in decimal pixel units using an ECC (Enhanced Correlation Coefficient) based algorithm. The method of claim 1,
Wherein the plurality of frame images include images acquired by a thermal imaging camera. The method of claim 1,
Wherein the object comprises a ladle used in a steel process. A high-resolution image acquisition apparatus for generating a high-resolution image using a plurality of low-resolution images,
An image capturing unit capturing a moving object to acquire a plurality of frame images,
A motion estimator for estimating a coordinate value in decimal pixel units of the relative motion of the current image with respect to the previous image using the plurality of frame images,
Selecting one unit pixel on the basis of the number of images corresponding to each coordinate value displayed in the plurality of unit pixels having the coordinate values of the estimated images as the center point and selecting one unit pixel among the images corresponding to the coordinate values in the selected unit pixel An image selection unit for selecting an image to which a super resolution technique is applied, and
A high-resolution image generating unit for generating a high-resolution image from the selected image using the super resolution technique,
Resolution image. delete The method of claim 9,
The image selection unit divides the selected unit pixel into a plurality of blocks and obtains a high resolution image for selecting an image having a coordinate value closest to the center of each block in which a video exists among the plurality of blocks as an image to which the super resolution technique is applied Device. The method of claim 9,
Wherein the image selecting unit selects a unit pixel having the largest number of images among the plurality of unit pixels. The method of claim 9,
Wherein the image capturing unit includes a thermal imaging camera. The method of claim 9,
Wherein the object comprises a ladle used in a steel process.

Images