KR102176511B1 - Apparatus and Method for processing image - Google Patents

Abstract

An image processing apparatus and an image processing method thereof are provided. The image processing apparatus includes a ray generator that generates light rays, a plurality of tree search units that perform a tree search on the generated rays, and a ray generated based on the number of tree search times for a previous image frame into a plurality of sub-regions. And a light ray allocating unit that allocates each of a region dividing unit to be divided and a plurality of subregions divided by the region dividing unit to a plurality of tree search units.

Description

Image processing apparatus and its image processing method {Apparatus and Method for processing image}

The present invention relates to an image processing apparatus and an image processing method thereof, and more particularly, to an image processing apparatus capable of rendering an image through a ray tracing technique, and an image processing method thereof.

Conventional graphics rendering work is being processed by a GPU (Graphics Processing Unit), but it is impossible to render realistic and three-dimensional realistic rendering. In recent years, realistic rendering techniques such as ray casting, ray tracing, and global illumination techniques have been developed as more realistic and three-dimensional rendering algorithms.

As shown in FIG. 1, the conventional image processing apparatus 100 includes a ray generator 110, a tree search unit 120, a cross check unit 130, and a shading unit 140. In this case, the operation load of the tree search unit 120 is the largest, and the performance of the tree search unit 120 has the greatest effect on the overall performance of the image processing apparatus 100. Accordingly, the image processing apparatus 100 includes a plurality of tree search units 120-1, 120-2, 120-3, .., as shown in FIG. 1 to improve performance.

The rays generated by the ray generator 110 are assigned to one of the plurality of tree search units 120-1, 120-2, 120-3,..., and the tree search unit ( 120) affects the cache memory hit rate. Since this greatly affects the performance of the tree search unit 120, a method of efficiently allocating rays to the tree search unit 120 is very important.

A conventional method of allocating light rays to a plurality of tree search units 120-1, 120-2, 120-3, ... is as shown in FIGS. 2A and 2B. Specifically, a plurality of light rays generated by the light ray generation unit 110 are FIFOs (First-In, First-) in a plurality of tree search units 120-1, 120-2, 120-3, ..., as shown in FIG. 2A. Out) method. That is, by checking the state of the input FIFOs of the plurality of tree search units 120-1, 120-2, 120-3, ..., an empty tree search unit is searched, and rays are sequentially allocated.

In addition, the plurality of rays generated by the ray generator 110 are divided into groups by the number corresponding to the plurality of tree search units 120-1, 120-2, 120-3, ..., as shown in FIG. 2B, Each group of rays is assigned to a corresponding tree search unit.

However, the method as described above has the following problems. First, in the method of allocating light rays in the FIFO method as shown in FIG. 2A, the temporal locality effect cannot be seen in the cache memory because the correlation between the rays processed by the tree search unit is low. Deterioration will occur.

In addition, the method of allocating rays to groups as shown in FIG. 2B can obtain a temporal locality effect and thus a higher cache memory hit ratio. However, when rays that require a lot of tree search are allocated to some groups, a load balancing problem occurs. Accordingly, a problem occurs in that the tree search unit 120 is not efficiently operated.

The present invention was conceived to solve the above-described problem, and an object of the present invention is to allocate light rays generated in consideration of cache memory efficiency and load balancing to a plurality of tree search units, thereby improving the performance of the tree search unit. It is to provide an image processing apparatus and an image processing method thereof.

In accordance with an embodiment of the present invention for achieving the above object, an image processing apparatus includes: a light beam generator configured to generate a light beam; A plurality of tree search units for performing a tree search on the generated rays; A region dividing unit for dividing the generated ray into a plurality of sub-regions based on tree search number information for a previous image frame; And a light ray allocating unit that allocates each of the plurality of subregions divided by the region dividing unit to the plurality of tree search units.

In addition, the region dividing unit stores a plurality of divided categories in which sub-regions are divided in a plurality of ways, and determines the division category in which the tree search time for the previous image frame is the shortest among the plurality of divided categories. It can be divided into a plurality of sub-areas to correspond to the divided category.

In addition, the region dividing unit divides the previous image frame into a plurality of sub-regions according to a division category, and determines the number of tree searches corresponding to pixels included in each of the plurality of sub-regions and the size of each of the plurality of sub-regions. A tree search value may be calculated by adding a value multiplied by a corresponding weight, and a tree search time for a split category may be determined using the calculated tree search value.

In addition, the region dividing unit calculates the tree search value for each of the plurality of divided categories, and converts the divided category having the smallest tree search value among the calculated plurality of tree search values into a divided category having the shortest tree search time. I can judge.

In addition, the region dividing unit may store information on the determined division category in a memory, and provide the ray allocator with information on the division category stored in the memory when sub-regions are divided for a next image frame.

In addition, the ray allocating unit additionally allocates a ray included in another sub-region for which the tree search is not completed to a tree search unit in which the tree search for the ray included in the corresponding sub-region among the plurality of tree search units has been completed. can do.

Further, a cross check unit for performing a cross check on the ray on which the tree search has been performed; And a shading unit that performs a shading operation on the ray on which the cross check has been performed.

On the other hand, according to an embodiment of the present invention for achieving the above object, an image processing method includes the steps of: generating a light ray; Dividing the generated ray into a plurality of sub-regions based on the tree search number information for the previous image frame; And allocating each of a plurality of subregions divided by the region dividing unit to a plurality of tree search units, and performing a tree search on the generated light rays by the plurality of tree search units.

In the dividing step, a plurality of division categories in which the sub-regions are divided in a plurality of methods are stored, and among the plurality of division categories, a division category in which a tree search time for the previous image frame is the shortest time is determined, and the It may be divided into a plurality of sub-areas to correspond to the determined division category.

In addition, the dividing may include dividing the previous image frame into a plurality of sub-regions according to a division category, and the number of tree searches corresponding to pixels included in each of the plurality of sub-regions and a size of each of the plurality of sub-regions A tree search value may be calculated by adding a value multiplied by a weight corresponding to, and a tree search time for a split category may be determined using the calculated tree search value.

In the dividing step, the tree search value is calculated for each of the plurality of divided categories, and the divided category having the smallest tree search value among the calculated plurality of tree search values is a divided category whose tree search time is the shortest time. It can be judged as.

In addition, the dividing may include storing information on the determined division category in a memory; And providing the light beam allocator with information on a division category stored in the memory when sub-regions are divided for the next image frame.

In the allocating step, the tree search unit in which the tree search for the light rays included in the corresponding sub-region among the plurality of tree search units has been completed is additionally added to the light rays included in the other sub-regions for which the tree search is not completed. Can be assigned.

Further, performing a cross check for the ray on which the tree search has been performed; And performing a shading operation on the ray on which the crossover inspection has been performed.

According to the various embodiments of the present invention as described above, by allocating a group of rays having high correlation to one tree search unit, it is possible to improve a cache memory hit ratio of the tree search unit through a temporal locality effect. Accordingly, there is an effect of improving the performance of the tree search unit and the overall performance of the image processing apparatus, and reducing the external memory bandwidth. In addition, by adaptively dividing the ray group into subgroups in consideration of the previous frame and assigning it to the tree search unit, it is possible to maintain balanced load balancing.

1 is a block diagram showing the configuration of a conventional image processing apparatus.
2A and 2B are diagrams for explaining a conventional ray allocation method;
3 is a block diagram showing the configuration of an image processing apparatus according to an embodiment of the present invention;
4 is a diagram showing a plurality of division categories according to an embodiment of the present invention;
5 is a diagram for describing a method of calculating a tree search value for each of a plurality of divided categories for an image frame according to an embodiment of the present invention; and
6 is a flowchart illustrating an image processing method according to an embodiment of the present invention.

Since these embodiments can apply various transformations and have various embodiments, specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the scope of the specific embodiment, it is to be understood to include all conversions, equivalents, or substitutes included in the disclosed spirit and technical scope. In describing the embodiments, when it is determined that a detailed description of a related known technology may obscure the subject matter, the detailed description thereof will be omitted.

Terms such as first and second may be used to describe various components, but the components should not be limited by terms. The terms are only used for the purpose of distinguishing one component from another component.

The terms used in the present application are used only to describe specific embodiments, and are not intended to limit the scope of the rights. Singular expressions include plural expressions unless the context clearly indicates otherwise. In the present application, terms such as "comprise" or "comprise" are intended to designate the existence of features, numbers, steps, actions, components, parts, or a combination thereof described in the specification, but one or more other It is to be understood that the presence or addition of features, numbers, steps, actions, components, parts, or combinations thereof, does not preclude in advance the possibility of being excluded.

In an embodiment, the'module' or'unit' performs at least one function or operation, and may be implemented as hardware or software, or a combination of hardware and software. In addition, a plurality of'modules' or a plurality of'units' are integrated into at least one module except for'modules' or'units' that need to be implemented with specific hardware and implemented as at least one processor (not shown). Can be.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings, and in the description with reference to the accompanying drawings, the same or corresponding constituent elements are assigned the same reference numerals, and redundant descriptions thereof will be omitted.

3 is a block diagram showing the configuration of an image processing apparatus 300 according to an embodiment of the present invention. As shown in FIG. 3, the image processing apparatus 300 includes a light ray generator 310, a light ray allocating unit 320, a region dividing unit 330, a memory 340, and a plurality of tree search units 350-1, 350. -2,350-3,..), a cross inspection unit 360 and a shading unit 370 are included. In this case, the image processing device 300 may be implemented as various image processing devices such as a smart phone, a smart TV, a tablet PC, and a notebook PC.

The ray generator 310 generates a ray for outputting image data according to a ray tracing technique. In this case, the light ray generator 310 may generate a light ray for each of a plurality of pixels constituting the image data.

Specifically, the light ray generator 310 may generate at least one ray based on the light ray generation information. In particular, the ray generator 310 may acquire a screen coordinate value of the ray based on the ray generation information. In addition, the ray generator 310 may generate a ray based on the obtained screen coordinate value and the camera position (ie, the position of the eye).

The ray allocating unit 320 includes a plurality of tree search units 350-1,350-2,350-3, based on subregion division information provided from the region division unit 330 on the rays generated by the ray generation unit 310. .) Can be assigned to each. In this case, the subregion division information may include information on a plurality of subregions that divide the image frame. That is, the ray allocating unit 320 uses the plurality of tree search units 350-1, 350-2, 350-3, ...) to each of the rays included in the sub-region adaptively divided for each frame by the region dividing unit 330. Can be assigned to

In addition, the ray allocating unit 320 terminates the tree search in the tree search unit in which the tree search for the rays included in the corresponding subregions among the plurality of tree search units 350-1,350-2,350-3,... It is possible to additionally allocate light rays included in other subregions that are not. For example, a tree search for a ray included in the subregion allocated to the first tree search unit 350-1 is terminated, and a ray included in the subregion allocated to the second tree search unit 350-2 When the tree search for is not finished, the ray allocating unit 320 transfers some of the rays included in the subregion allocated to the second tree search unit 350-2 to the first tree search unit 350-1. Can be reassigned. In this way, the ray allocating unit 320 may adaptively allocate the ray of which the tree search has not been completed to the tree search unit on which the tree search has been completed, so that a faster tree search may be possible.

On the other hand, since the region dividing unit 330 divides the sub-region based on the tree search number information for the previous frame, the ray allocating unit 320 uses a plurality of tree search units in a preset manner for the ray included in the first frame. It can be assigned as (350-1,350-2,350-3,..).

The region dividing unit 330 divides the next frame into a plurality of sub-regions based on the tree search number information for the previous image frame.

Specifically, the region dividing unit 330 may store a plurality of division categories in which sub-regions are divided in a plurality of ways. For example, when an image frame is composed of 8X8 and is divided into four sub-regions, as shown in FIG. 4, the region dividing unit 330 includes first to ninth sub-categories 410 to 490. ) Can be saved. Specifically, in the first division category 410, the first subregion at the upper left has a size of 2X2, the second subregion at the upper right has a size of 2X6, and the third subregion at the lower left has a size of 6X2. And the fourth sub-region at the bottom right has a size of 6X6. In the second divided category 420, the first subregion at the upper left has a size of 2X4, the second subregion at the upper right has a size of 2X4, and the third subregion at the lower left has a size of 6X4, The fourth sub-area at the bottom right has a size of 6X4. In the third divided category 430, the first sub-region at the upper left has a size of 2X6, the second sub-region at the upper right has a size of 2X2, and the third sub-region at the lower left has a size of 6X6, The fourth sub-area at the bottom right has a size of 6X2. In the fourth divided category 440, the first sub-region at the upper left has a size of 4X2, the second sub-region at the upper right has a size of 4X6, and the third sub-region at the lower left has a size of 4X2, The fourth sub-area at the bottom right has a size of 4X6. The fifth division category 450 has a size of 4X4 for all of the first to fourth sub-regions. In the sixth divided category 460, the first sub-region at the upper left has a size of 4X6, the second sub-region at the upper right has a size of 4X2, and the third sub-region at the lower left has a size of 4X6, The fourth sub-area at the bottom right has a size of 4X2. In the seventh division category 470, the first sub-region at the upper left has a size of 6X2, the second sub-region at the upper right has a size of 6X6, and the third sub-region at the lower left has a size of 2X2, The fourth sub-area at the bottom right has a size of 2X6. In the eighth division category 480, the first subregion at the upper left has a size of 6X4, the second subregion at the upper right has a size of 6X4, and the third subregion at the lower left has a size of 2X4, The fourth sub-area at the bottom right has a size of 2X4. In the ninth divided category 490, the first sub-region at the upper left has a size of 6X6, the second subregion at the upper right has a size of 6X2, and the third sub-region at the lower left has a size of 2X6, The fourth sub-area at the bottom right has a size of 2X2.

Meanwhile, as described above, it is only an exemplary embodiment that the division category is divided into four sub-regions, and may be divided into a plurality of sub-regions other than four. In this case, the number of subregions of the divided category may correspond to the number of tree search units 350. For example, when the number of tree search units 350 is eight, the number of sub-areas of the divided category may also be eight.

In addition, the region dividing unit 330 determines a divided category in which the tree search time for the previous image frame is the shortest among the stored divided categories 410 to 490, and divides it into a plurality of sub-regions to correspond to the determined divided category. can do. Specifically, the region dividing unit 330 may calculate a tree search value proportional to the tree search time, and determine a tree search time for each of the plurality of divided categories using the calculated tree search value.

At this time, the region dividing unit 330 divides the previous image frame into a plurality of subregions according to a specific segmentation category, and the number of tree searches corresponding to pixels included in each of the plurality of subregions and the size of each of the plurality of subregions A tree search value can be calculated by adding a value multiplied by a corresponding weight. The tree search value (P _cat ) can be calculated by Equation 1 below.

Here, TRV_CRT(i) is the number of tree searches of the i-th pixel, w1 is the weight of the first sub-region, w2 is the weight of the second sub-region, w3 is the weight of the third sub-region, and w4 is the weight of the fourth sub-region. Weight. In this case, each of w1, w2, w3, and w4 corresponds to the size of the corresponding subregion.

For example, when the number of tree search times of each pixel included in the previous image frame is as shown in FIG. 5, the region dividing unit 330 includes a plurality of division categories 410 to 490 as shown in Equation 2 below. You can calculate the tree search value for each.

In addition, the region dividing unit 330 may determine a division category having the smallest tree search value among the tree search values calculated for each of the plurality of division categories 410 to 490 as a division category having the shortest tree search time. have. For example, as described above, when a tree search value for each of the first to ninth subcategories 410 to 490 is calculated, the region divider 330 is divided into a sixth division with the smallest tree search value. The category can be determined as a divided category in which the tree search time is the shortest time.

In addition, the region dividing unit 330 may provide information on a division category in which the tree search time for the previous frame image is the shortest time to the ray allocating unit 320 as sub-region division information for the next image frame. For example, as calculated in Equation 2, when it is determined that the segmentation category having the shortest tree search time is the sixth segmentation category 460, the region segmentation unit 330 is assigned to the sixth segmentation category 460. Information about the light may be provided to the light ray allocating unit 320.

The light ray allocating unit 320 may allocate the light rays to the plurality of tree search units 350-1, 350-2, 350-3, ... based on the received sub-region division information. For example, when the ray allocating unit 320 receives information on the sixth division category 460 as sub-region division information, the ray allocator 320 is generated based on the sixth division category 460 Each of the rays may be provided to the plurality of tree search units 350-1, 350-2, 350-3, .. Specifically, the ray allocating unit 320 allocates the rays included in the first subregion of the sixth division category 460 to the first tree search unit 350-1, and 2 The light rays included in the sub-region are allocated to the second tree search unit 350-2, and the rays included in the third sub-region of the sixth division category 460 are assigned to the third tree search unit 350-3. After allocating, the rays included in the fourth subregion of the sixth division category 460 may be allocated to the fourth tree search unit 350-4.

In addition, the region dividing unit 330 stores information on the division category determined for the previous frame in the memory 340, and stores information on the division category stored in the memory 340 when sub-regions are divided for the next image frame. A light ray allocating unit 320 may be provided.

The tree search unit 350 performs a tree search on the received ray. At this time, the tree search unit 350 may use a kd-tree, which is a kind of spatial partitioning tree method. The kd-tree includes a box node, an inner node, and a leaf node, and the leaf node includes a triangle list for pointing at least one triangle information included in the geometric data. I can. In an embodiment of the present invention, when triangle information included in geometric data is implemented as an array, a triangle list included in the leaf node 930 may correspond to an array index.

However, as described above, the use of kd-tree is only an example, and various types of tree search techniques such as Bounding Volume Hierarchy (BVH) may be performed.

In particular, since the image processing apparatus 300 may generate a lot of load in the tree search, as shown in FIG. 3, a plurality of tree search units is provided with a plurality of tree search units 350-1,350-2,350-3. (350-1,350-2,350-3) Each can independently perform a tree search for a ray.

The cross check unit 360 performs cross check on the ray on which the tree search has been performed. Specifically, the intersection inspection unit 360 searches for a leaf node that intersects the ray. In addition, the intersection inspection unit 360 reads a list of triangles included in the intersecting leaf nodes. The cross check unit 360 may read coordinate information of a triangle list and perform a cross test on a given ray. In addition, the intersection inspection unit 360 calculates the coordinate value of the ray-triangle hit point by using the distance from the triangle hit by the ray and the vector value of the given ray.

The shading unit 370 performs a shading operation on the ray on which the cross check has been performed. Specifically, the shading unit 370 may calculate a color value of a ray-triangle hit point by using information stored in a material memory and a triangle information cache. In addition, the shading unit 370 may perform a shading operation based on the calculated color value.

Hereinafter, an image processing method of an image processing apparatus will be described with reference to FIG. 6.

First, the image processing apparatus 300 generates light rays (S610). In this case, the image processing apparatus 300 may generate light rays for each pixel of the image frame.

Then, the image processing apparatus 300 divides the light rays generated based on the tree search number information for the previous frame into a plurality of sub-regions (S620). Specifically, the image processing apparatus 300 stores a plurality of divided categories in which sub-regions are divided in a plurality of ways, and determines and determines a divided category in which the tree search time for the previous image frame is the shortest among the plurality of divided categories. It can be divided into a plurality of sub-areas to correspond to the divided category. More specifically, the image processing apparatus 300 divides the previous image frame into a plurality of sub-regions according to a specific division category, and the number of tree searches corresponding to pixels included in each of the plurality of sub-regions and each of the plurality of sub-regions A tree search value can be calculated by adding a value multiplied by a weight corresponding to the size. In addition, the image processing apparatus 300 calculates a tree search value for each of the plurality of divided categories, and determines the divided category having the smallest tree search value among the plurality of tree search values calculated as the divided category having the shortest tree search time. can do.

In addition, the image processing apparatus 300 allocates a plurality of sub-regions to a plurality of tree search units 350-1, 350-2, 350-3, .. (S630). Specifically, the image processing apparatus 300 divides a plurality of sub-areas based on information on a divided category in which the tree search time is the shortest time, and allocates rays included in each of the plurality of sub-areas to a corresponding tree search unit. can do.

In addition, the image processing apparatus 300 performs a tree search on the rays included in the allocated sub-region (S640). In this case, the image processing apparatus 300 may independently perform a tree search for a ray using a plurality of tree search units 350-1, 350-2, 350-3, ...

In addition, the image processing apparatus 300 performs a cross-search and shading operation on the tree-searched ray (S650 and S660).

As described above, by allocating a ray group having high correlation to one tree search unit for each sub-region, it is possible to improve the hit rate of the cache memory of the tree search unit. Accordingly, there is an effect of improving the performance of the tree search unit and the overall performance of the image processing apparatus, and reducing the external memory bandwidth. In addition, by adaptively dividing the ray group into subgroups in consideration of the previous frame and assigning it to the tree search unit, it is possible to maintain balanced load balancing.

The device according to the present embodiments includes a processor, a memory for storing and executing program data, a permanent storage such as a disk drive, a communication port for communicating with an external device, a touch panel, a key, a button, etc. It may include a user interface device and the like. Methods implemented as software modules or algorithms may be stored on a computer-readable recording medium as computer-readable codes or program instructions executable on the processor. Here, as a computer-readable recording medium, a magnetic storage medium (e.g., read-only memory (ROM), random-access memory (RAM), floppy disk, hard disk, etc.) and optical reading medium (e.g., CD-ROM ) And DVD (Digital Versatile Disc). The computer-readable recording medium is distributed over network-connected computer systems, so that computer-readable codes can be stored and executed in a distributed manner. The medium is readable by a computer, stored in memory, and executed on a processor.

This embodiment may be represented by functional block configurations and various processing steps. These functional blocks may be implemented with various numbers of hardware or/and software configurations that perform specific functions. For example, the embodiment is an integrated circuit configuration such as memory, processing, logic, a look-up table, etc., capable of executing various functions by controlling one or more microprocessors or other control devices Can be hired. Similar to how components can be implemented with software programming or software elements, this embodiment includes various algorithms implemented with a combination of data structures, processes, routines or other programming components, including C, C++, Java ( Java), an assembler, etc. may be implemented in a programming or scripting language. Functional aspects can be implemented with an algorithm running on one or more processors. In addition, the present embodiment may employ a conventional technique for electronic environment setting, signal processing, and/or data processing. Terms such as “mechanism”, “element”, “means”, and “configuration” can be used broadly and are not limited to mechanical and physical configurations, etc. The terms are a series of routines of software in connection with a processor, etc. Can include the meaning of ).

The specific implementations described in the present embodiment are examples, and do not limit the technical scope in any way. For brevity of the specification, descriptions of conventional electronic configurations, control systems, software, and other functional aspects of the systems may be omitted. In addition, the connection or connection members of the lines between the components shown in the drawings exemplarily represent functional connections and/or physical or circuit connections, and in an actual device, various functional connections that can be replaced or additionally It may be referred to as a connection, or circuit connections.

In the present specification (especially the claims), the use of the term "above" and a reference term similar thereto may correspond to both the singular and the plural. In addition, when a range is described, it includes individual values belonging to the above range (unless otherwise stated), it is the same as describing each individual value constituting the range in the detailed description. Finally, if there is no explicit order or contradictory description of the steps constituting the method, the steps may be performed in an appropriate order. It is not necessarily limited to the order of description of the steps. The use of all examples or illustrative terms (for example, etc.) is merely for describing the technical idea in detail, and the scope is not limited due to the examples or illustrative terms unless limited by the claims. In addition, those skilled in the art can recognize that various modifications, combinations, and changes may be configured according to design conditions and factors within the scope of the appended claims or their equivalents.

310: ray generation unit 320: ray allocation unit
330: area division unit 340: memory
350: tree search unit 360: cross check unit
370: shading unit

Claims (14)

A light ray generator that generates light rays;
A plurality of tree search units for performing a tree search on the generated rays;
A region dividing unit for dividing the generated ray into a plurality of sub-regions based on tree search number information for a previous image frame; And
And a light beam allocating unit that allocates each of the plurality of sub-regions divided by the region dividing unit to the plurality of tree search units. The method of claim 1,
The area division unit,
A plurality of divided categories in which the sub-regions are divided in a plurality of methods are stored, and a plurality of divided categories are determined to correspond to the determined divided categories by determining a divided category whose tree search time for the previous image frame is the shortest among the plurality of divided categories. The image processing apparatus, characterized in that divided into sub-regions of. The method of claim 2,
The area division unit,
The previous image frame is divided into a plurality of sub-regions according to a division category, and a value obtained by multiplying the number of tree searches corresponding to pixels included in each of the plurality of sub-regions and a weight corresponding to the size of each of the plurality of sub-regions is obtained. In addition, a tree search value is calculated, and a tree search time for a divided category is determined by using the calculated tree search value. The method of claim 2,
The area division unit,
An image processing method characterized in that, by calculating a tree search value for each of the plurality of divided categories, the divided category having the smallest tree search value among the calculated plurality of tree search values is determined as the divided category having the shortest tree search time. Device. The method of claim 2,
The area division unit,
And storing information on the determined segmentation category in a memory, and providing information on the segmentation category stored in the memory to the ray allocating unit when subregions are segmented for a next image frame. The method of claim 1,
The light beam allocating unit,
An image processing, comprising: additionally allocating a light ray included in another sub-region for which the tree search has not been completed, to a tree search unit in which a tree search for a ray included in a corresponding sub-region among the plurality of tree search units has been completed. Device. The method of claim 1,
An intersection inspection unit for performing an intersection inspection on the ray on which the tree search has been performed; And
And a shading unit configured to perform a shading operation on the ray on which the cross-check has been performed. Generating a light beam by a light ray generator;
Dividing, by a region dividing unit, the generated ray into a plurality of subregions based on the tree search number information for the previous image frame;
Allocating, by a light ray allocating unit, each of a plurality of sub-regions divided by the region dividing unit to a plurality of tree search units;
And performing a tree search on the generated rays by the plurality of tree search units. The method of claim 8,
The dividing step,
A plurality of divided categories in which the sub-regions are divided in a plurality of methods are stored, and a plurality of divided categories are determined to correspond to the determined divided categories by determining a divided category whose tree search time for the previous image frame is the shortest among the plurality of divided categories. The image processing method, characterized in that divided into sub-regions of. The method of claim 9,
The dividing step,
The previous image frame is divided into a plurality of sub-regions according to a division category, and a value obtained by multiplying the number of tree searches corresponding to pixels included in each of the plurality of sub-regions and a weight corresponding to the size of each of the plurality of sub-regions is obtained. In addition, a tree search value is calculated, and a tree search time for a divided category is determined by using the calculated tree search value. The method of claim 9,
The dividing step,
An image processing method characterized in that, by calculating a tree search value for each of the plurality of divided categories, the divided category having the smallest tree search value among the calculated plurality of tree search values is determined as the divided category having the shortest tree search time. Way. The method of claim 9,
The dividing step,
Storing information on the determined division category in a memory; And
And providing information on a segmentation category stored in the memory to the light beam allocating unit when segmenting a subregion for a next image frame. The method of claim 8,
The allocating step,
An image processing, comprising: additionally allocating a light ray included in another sub-region for which the tree search has not been completed, to a tree search unit in which a tree search for a ray included in a corresponding sub-region among the plurality of tree search units has been completed. Way. The method of claim 8,
Performing a cross check on the ray on which the tree search has been performed; And
And performing a shading operation on the ray on which the cross check has been performed.

Images