KR102213147B1 - Image Processing Method and Vehicle supporting the same - Google Patents

Abstract

The present invention sets a threshold area and a threshold value for determining the complexity of a screen based on a far-infrared sensor for collecting a front image of a vehicle and encoding information of a first frame constituting the image, and The present invention relates to an image processing method including a control unit for controlling to determine a size of a target block to be designated as a coding unit using the threshold region and the threshold value when intra prediction encoding two frames in a screen, and a vehicle supporting the same.

Description

Image Processing Method and Vehicle supporting the same}

The present invention relates to image processing, and more particularly, to an image processing method supporting the determination of a prediction mode of intra prediction encoding in a process of H.264 encoding of a far-infrared image, and a vehicle supporting the same.

An encoder conforming to the H.264 standard may perform intra-prediction encoding, that is, intra-mode encoding, in consideration of the direction of a block by reflecting local characteristics in a spatial domain of a screen represented by an image. When a block or a macroblock (MB) is encoded in the intra mode, the target block to be predicted may be configured by using an already reconstructed block located adjacent to the target block.

Meanwhile, an encoder conforming to the H.264 standard may perform various prediction modes for encoding an image. In this regard, an encoder conforming to the H.264 standard may perform predictive encoding on luminance blocks in units of 4x4 subblocks or 16x16 macroblocks. In this case, an encoder conforming to the H.264 standard can perform 9 prediction modes according to directionality when predictive encoding for luminance blocks is performed in units of 4x4 sub-blocks, and directionality is performed in units of 16x16 macroblocks. According to, four prediction modes can be performed. In addition, macroblocks for prediction in a spatial domain may be classified into I16MB or I4MB, and accordingly, different prediction modes may be performed. For example, in the case of I16MB, a 16x16 macroblock is predicted in four intra-space prediction directions, and in the case of I4MB, a prediction error can be calculated by performing prediction in nine intra-space prediction directions in a unit of 4x4 size. .

In this regard, in the prior art, an optimal prediction mode may be determined using a rate-distortion optimization (RDO) technique. The RDO technique is a method of determining an optimal prediction mode after encoding in all possible prediction modes. Therefore, although the RDO technique can increase the coding efficiency through the optimal prediction mode, it takes a considerable amount of time in the process of determining the optimal prediction mode.

An object of the present invention to solve such a conventional problem is to determine a prediction mode for intra prediction encoding according to the statistical characteristics of the image when encoding a far-infrared image in front of a vehicle according to the H.264 standard. It is to provide an image processing method that can improve processing speed while complying with standards and a vehicle supporting the same.

The vehicle of the present invention sets a threshold area and a threshold value for determining the complexity of a screen based on a far-infrared sensor for collecting the front image of the vehicle and encoding information of a first frame constituting the image, and A control unit for controlling the size of a target block to be designated as a coding unit is determined by using the threshold region and the threshold value during intra prediction encoding of the constituting second frame.

The control unit may control to designate the first frame a certain number of times at a certain time interval, and to designate a frame temporally located behind the first frame as the second frame.

The encoding information of the first frame is a size of blocks included in the first frame designated as the coding unit and pixels included in the blocks for reference pixels to which a reference block adjacent to the blocks abuts May contain the variance value of

The control unit sets the critical area as an area corresponding to a boundary between first blocks having a relatively large size of the blocks and second blocks having a relatively small size of the blocks, and the first A screen area in which blocks are located may be designated as an upper area, a screen area in which the first blocks are located below the critical area may be designated as a lower area, and a screen area excluding the upper area and the lower area may be designated as a middle area. have.

When the number of first blocks having the variance value as a specific value and the number of second blocks are the same, the controller may set the specific value as the threshold value.

The controller sets the size of the target block located in the upper area and the lower area to be relatively large, and when the variance value of the target block located in the middle area is less than the threshold value, the size of the target block is relatively It is set to be large, and when the variance value of the target block located in the middle area is greater than the threshold value, the size of the target block may be set relatively small.

The image processing method of the present invention includes the steps of collecting a front image of the vehicle, setting a threshold area and a threshold value for determining the complexity of a screen based on encoding information of a first frame constituting the image, and The second frame constituting the image may be subjected to intra prediction encoding, controlling to determine a size of a target block to be designated as a coding unit using the threshold region and the threshold value.

The step of setting the threshold area and the threshold value includes designating the first frame a certain number of times at a certain time interval, and controlling a frame located temporally behind the first frame to designate the second frame. Can include.

In the setting of the threshold region and the threshold value, the sizes of blocks included in the first frame designated as the coding unit and the blocks for reference pixels to which a reference block adjacent to the blocks contact each other may be performed. And including the variance values of the included pixels in the encoding information of the first frame.

The setting of the threshold area and the threshold value may include setting an area corresponding to a boundary of first blocks having a relatively large size of the blocks and second blocks having a relatively small size of the blocks as the critical area, Designating a screen area in which the first blocks are located at an upper end of the critical area as an upper area, designating a screen area in which the first blocks are located at a lower end of the critical area as a lower area, and the upper area and the It may include the step of designating the screen area excluding the lower area as the middle area.

The step of setting the threshold region and the threshold value may include setting the specific value as the threshold value when the number of the first blocks having the variance value as a specific value and the number of the second blocks are the same. have.

The controlling to determine the size of the target block may include setting the size of the target block located in the upper region and the lower region to be relatively large, and the variance value of the target block located in the middle region is greater than the threshold value. If it is small, setting the size of the target block to be relatively large, and when the variance value of the target block located in the middle region is greater than the threshold value, setting the size of the target block to be relatively small. have.

According to the image processing method of the present invention and a vehicle supporting the same, the present invention can improve the encoding speed by determining a prediction mode for intra prediction encoding early, and by determining the prediction mode according to the statistical characteristics of the image. It can reduce the drop in efficiency.

1 is a diagram schematically showing a configuration of a vehicle supporting an image processing method according to an exemplary embodiment of the present invention.
FIG. 2 is a diagram illustrating a prediction mode performed in units of 16x16 macroblocks among prediction modes for intra prediction encoding according to the present invention.
FIG. 3 is a diagram illustrating a prediction mode performed in units of 4x4 subblocks among prediction modes for intra prediction encoding according to the present invention.
4 is a diagram showing in more detail a prediction mode determination unit among the configuration of a control unit of a vehicle according to the present invention.
5 is a diagram for explaining a critical area setting according to the present invention.
6 is a diagram for describing a threshold value setting according to the present invention.
7 is a diagram for describing an image processing method based on a threshold region and a threshold value according to the present invention.
8 is a diagram for explaining the effect of the image processing method of the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in more detail with reference to the accompanying drawings. However, in describing the embodiments, descriptions of technical contents that are well known in the technical field to which the present invention pertains and are not directly related to the present invention will be omitted as much as possible. This is to more clearly convey the core of the present invention by omitting unnecessary description.

1 is a diagram schematically showing a configuration of a vehicle supporting an image processing method according to an embodiment of the present invention, and FIG. 2 is a prediction mode performed in units of 16x16 macroblocks among prediction modes for intra prediction encoding according to the present invention. It is a figure for explaining. 3 is a diagram for explaining a prediction mode performed in units of 4x4 subblocks among prediction modes for intra prediction coding according to the present invention.

First, referring to FIG. 1, the vehicle 100 of the present invention may include a far infrared ray sensor 110 and a controller 160. Vehicle 100 having such a configuration may encode the far-infrared image collected by the far-infrared sensor 110 according to the H.264 standard through the H.264 encoder 60 included in the controller 160. In this case, in the encoding process, the vehicle 100 may determine a prediction mode of intra prediction encoding according to the statistical characteristics of the image through the prediction mode determiner 80 included in the controller 160.

The far-infrared ray sensor 110 may detect a far-infrared ray radiated from an object and obtain a far-infrared ray image based on the detected far-infrared ray. In addition, the far-infrared ray sensor 110 may transmit the obtained far-infrared ray image to the controller 160. The far-infrared ray sensor 110 may be disposed in front of the vehicle 100 to support collection of a far-infrared ray image in front of the vehicle. The far-infrared image acquired through the far-infrared sensor 110 may be a video including only a luminance component composed of a screen of several tens of frames per second.

The controller 160 may include an H.264 encoder 60 and a prediction mode determiner 80 to encode a far infrared image transmitted through the far infrared sensor 110 according to the H.264 standard. The controller 160 may encode a far-infrared image in front of the vehicle including only the luminance component through the H.264 encoder 60, and determine a prediction mode performed in the encoding process through the prediction mode determiner 80.

The encoding process of the far-infrared image according to the H.264 standard includes intra prediction encoding, that is, intra encoding. Intra encoding may be a method of encoding an image corresponding to one frame when encoding a video composed of a plurality of frames. For example, intra-encoding may be intra-prediction encoding in consideration of a direction of a macroblock by reflecting a local characteristic in a spatial region of a screen representing a corresponding image. In this regard, the macroblock may be a prediction unit in which a plurality of pixels are grouped for motion compensation or prediction in the video encoding method of the H.264 standard. That is, the process of encoding an image may be performed in units of macroblocks. Also, the macro block can be divided into sub-blocks. For example, a 16x16 macroblock can be divided into four 4x4 subblocks.

Regarding the above-described intra encoding, the H.264 encoder 60 may perform predictive encoding on luminance blocks in units of 4x4 subblocks or 16x16 macroblocks. In addition, the H.264 encoder 60 may perform various prediction modes to encode an image. Regarding the above-described prediction mode, the H.264 encoder 60 may designate an already reconstructed block adjacent to the target block to be predicted as the reference block. In this case, the H.264 encoder 60 selects pixels (hereinafter referred to as reference pixels) from among the pixels included in the reference block, and is included in the target block based on the pixel values of the reference pixels. The pixel value of the pixels can be predicted.

In this regard, as shown in FIG. 2, when the H.264 encoder 60 performs predictive encoding on luminance blocks in units of 16x16 macroblocks, the H.264 encoder 60 designates a macroblock located at the top or left of the target block as a reference block. And four prediction modes may be performed according to the direction. For example, in prediction mode 0, a macroblock positioned above the target block to be predicted may be designated as a reference block and may be predicted in the vertical direction. In addition, in prediction mode 1, a macroblock located to the left of a target block may be designated as a reference block to predict in a horizontal direction. In addition, in prediction mode 2, macroblocks located above and to the left of the target block may be designated as reference blocks and may be predicted based on an average value of corresponding pixel values. Finally, in prediction mode 3, macroblocks located at the top and left of the target block may be designated as reference blocks and interpolated in a diagonal direction for prediction.

In addition, as shown in FIG. 3, when predictive encoding for luminance blocks is performed in units of 4x4 sub-blocks, the H.264 encoder 60 is a reference block at the top, left, top left, or right top of the target block. It can be designated as a located macroblock, and 9 prediction modes can be performed according to the direction. For example, the prediction mode is prediction mode 0 in the vertical direction (prediction mode 0), prediction mode 1 in the horizontal direction, prediction mode 2 based on the average value, and prediction mode 3 in the lower left diagonal direction. (prediction mode 3), prediction mode 4 in the lower right diagonal direction, prediction mode 5 in the vertical right direction, prediction mode 6 in the lower horizontal direction, Prediction mode 7 or prediction mode 8 in the upper horizontal direction may be included.

In the case of performing intra coding as described above, the number of prediction modes may be smaller and the processing speed may be relatively faster to perform in 16x16 macroblock units than in 4x4 subblock units. In addition, a far-infrared ray image in front of a vehicle may have a high probability of including information on the sky at the top of the screen, and a high probability of including information on a road at the bottom of the screen. Since the image is relatively uniformly expressed in the upper region corresponding to the sky and the lower region corresponding to the road, a prediction mode in units of 16x16 macroblocks may be selected. In addition, in the middle area of the screen where buildings, vehicles, or pedestrians are mainly expressed, a prediction mode (hereinafter referred to as 4x4 mode) is selected for a complex area according to the complexity of the image, and a relatively uniform area is 16x16. A prediction mode in units of macroblocks (hereinafter referred to as 16x16 mode) may be selected. However, it is not selected only in units of blocks described above, and a selection may be made consistent with the technical idea of the present invention. For example, according to the complexity of the image, a complex area can be predicted by setting the size of a target block smaller than that of a relatively non-complex area.

The prediction mode determiner 80 may function to perform encoding in a designated prediction mode in the process of the H.264 encoder 60 encoding an image. The prediction mode determiner 80 may collect encoding information of an image corresponding to a specific frame among far-infrared images in front of the vehicle from the H.264 encoder 60. Encoding information of an image corresponding to a specific frame may include information on a prediction mode used during encoding (eg, a size of a block designated as a coding unit). For example, it may include information on whether blocks corresponding to a specific area of a screen represented by an image are predicted in a 4x4 mode or a 16x16 mode. In addition, encoding information of an image corresponding to a specific frame may include variance values for reference pixels calculated when the prediction mode is performed.

The prediction mode determiner 80 may set a critical region based on the above-described prediction mode information. The critical region may be a reference region capable of classifying a screen corresponding to a specific frame into upper, middle, and lower regions. In this regard, the prediction mode determiner 80 may determine a distribution of the selected prediction modes (4x4 mode and 16x16 mode) in a process of encoding an image corresponding to a specific frame. Also, the prediction mode determiner 80 may determine blocks corresponding to a boundary between the 4x4 mode and the 16x16 mode based on the distribution of the selected prediction mode. The prediction mode determiner 80 may set an area in which blocks corresponding to the boundary are distributed as a critical area. The prediction mode determiner 80 may classify a screen corresponding to an image into upper, middle, and lower regions based on the critical region.

Also, the prediction mode determiner 80 may set a threshold value based on the above-described variance value. The threshold value may be a reference value for selecting a prediction mode of blocks included in the middle area of the screen. For example, the prediction mode determiner 80 may determine whether to predict a target block to be encoded in a 4x4 mode or a 16x16 mode based on a threshold value.

The prediction mode determiner 80 makes the H.264 encoder 60 predict the upper and lower regions of the screen in a 16x16 mode based on the set threshold region and the threshold value, and the middle region to predict the prediction mode determined based on the threshold value. Can function. Through the above-described function, it is possible to reduce the load of an operation for determining an optimal prediction mode in a process of the H.264 encoder 60 encoding an image.

FIG. 4 is a diagram illustrating in more detail a prediction mode determining unit among the configuration of a control unit of a vehicle of the present invention, and FIG. 5 is a diagram for explaining setting a critical region according to the present invention. And Figure 6 is a diagram for explaining the threshold value setting of the present invention.

First, referring to FIG. 4, the prediction mode determination unit 80 may include a threshold region setting unit 81 and a threshold value setting unit 83. The prediction mode determiner 80 is a threshold that is a criterion for classifying an area of a screen expressed by an image through the threshold area setting unit 81 based on the encoding information of the image corresponding to the specific frame collected from the H.264 encoder 60. An area may be set, and a threshold value that is a criterion for determining a prediction mode of blocks corresponding to the middle area of the screen may be set through the threshold value setting unit 83. In this regard, an image corresponding to a specific frame is a far-infrared image and may include only a luminance component. In addition, since the corresponding image corresponds to a specific frame of a moving picture composed of several tens of frames per second, it is necessary to reset the threshold region and the threshold value by a predetermined number of times every predetermined time. Accordingly, the prediction mode determiner 80 may collect encoding information of an image corresponding to a specific frame from the H.264 encoder 60 for a predetermined number of times every predetermined time. That is, the H.264 encoder 60 transmits the variance values of the reference pixels calculated in the process of determining the prediction mode by applying the RDO technique and information on the determined prediction mode to the prediction mode determiner 80 a certain number of times every predetermined time. I can. In addition, the set threshold region and threshold value may be used in a process in which the H.264 encoder 60 encodes a frame after a specific frame.

The critical region setting unit 81 may set a critical region capable of adaptively classifying a screen represented by an image corresponding to a specific frame into upper, lower, and middle regions. In this regard, referring to FIG. 5, the screen 501 may be a far-infrared image transmitted from the far-infrared sensor 110. The H.264 encoder 60 may determine a prediction mode by applying the RDO technique to the image of screen 501. The screen 503 may include information on a prediction mode determined by the RDO technique collected from the H.264 encoder 60. As shown in screen 503, a block with a wide interval may be a block predicted in a 16x16 mode, and a block with a narrow interval may be a block predicted in a 4x4 mode. In order to determine the upper area of the screen, the critical area setting unit 81 sets the coordinates of the block located at the top left of the screen at the top of the block predicted in the 16x16 mode at the top of the screen at coordinates A and the block at the top right The coordinates of may be designated as coordinates B, and the coordinates of the block located at the bottom of the center may be designated as coordinates. In this case, the critical area setting unit 81 can set an arbitrary quadratic polynomial passing through coordinates A, coordinates B, and coordinates C. When the quadratic polynomial is expressed as a curve in the coordinate area, the area through the curve is designated as a critical area. Can be set, and the upper part of the curve can be set as the upper area of the screen. In addition, the critical area setting unit 81 may designate a coordinate D, a coordinate E, and a coordinate F for setting the lower area of the screen in the same or similar to the above-described method. In this case, the critical area setting unit 81 may set the area where the curve corresponding to the second-order polynomial passes as the critical area by using an arbitrary quadratic polynomial passing through the specified coordinate D, the coordinate E, and the coordinate F. The lower part of the screen can be set as the lower part of the screen. Also, the critical area setting unit 81 may set an area excluding an upper area and a lower area of the screen as a middle area of the screen.

The threshold value setting unit 83 may adaptively set a threshold value that is a reference for determining a prediction mode of an intermediate region of a screen corresponding to a specific frame. The threshold value setting unit 83 may set a threshold value using variance values for reference pixels collected from the H.264 encoder 60. In this regard, referring to FIG. 6, a screen 601 represents a target block to be encoded and a reference block located adjacent to the target block. The H.264 encoder 60 converts a pixel in contact with the target block in an already coded macroblock (reference block) located at the top left of the target block to be coded to the pixel A, and the target block in the reference block located at the top of the target block. Abutting pixels may be designated as pixels B, and pixels in a reference block located to the left of the target block may be designated as pixels C. In this regard, the H.264 encoder 60 determines a prediction mode for a specific frame (eg, the first frame of a video collected from the far-infrared ray sensor 110) using the RDO method, and stores the coding unit of each macroblock and simultaneously Alternatively, variance values for pixels A, B, and C corresponding to a corresponding macroblock may be calculated and stored at predetermined time intervals. When only some of the pixels A, B, or C exist according to the position of the macroblock, the H.264 encoder 60 can calculate the variance value only for the pixels of the existing reference block, and the first macro of the frame If all of the pixels A, the pixels B, and the pixels C do not exist as in the block, the variance value may be set to 0 for processing. For example, the variance value of the macroblock corresponding to the first row and the first column of the frame can be set to 0, and the macroblocks included in the first row except for the macroblock corresponding to the first row and the first column of the frame are pixels. Macroblocks included in the first column excluding the variance value for C and the macroblocks corresponding to the first row and the first column may be designated as variance values for the pixels B. Macroblocks other than the above-described macroblocks may be designated as variance values for the pixels A, B, and C.

The threshold value setting unit 83 may set a threshold value based on encoding information of an image corresponding to a specific frame including the variance value calculated by the above-described method. In this regard, screen 603 is a screen in which the number of macroblocks per variance corresponding to the 4x4 mode and the 16x16 mode is displayed as a graph. The threshold value setting unit 83 may set a variance value indicated by an intersection point of the curve corresponding to the 4x4 mode and the curve corresponding to the 16x16 mode as the threshold value, as shown in the screen 603.

7 is a diagram for describing an image processing method based on a threshold region and a threshold value according to the present invention.

The far-infrared image in front of the vehicle may be an image containing only the luminance component, and if encoding is performed using the RDO technique, there is a high probability that information about the sky is included at the top of the screen, and the information about the road at the bottom of the screen is Since the probability of including information is high, a relatively uniform image appears in the corresponding region, and thus, a relatively small amount of computation can be predicted in a 16x16 mode. As described above, the H.264 encoder 60 may determine a prediction mode based on the threshold region and the threshold value by applying that the complexity of the target block to be encoded and the variance of reference pixels adjacent to the target block are correlated. have. The threshold region and the threshold value may be adaptively determined through the prediction mode determiner 80 according to an image.

In step S101, the H.264 encoder 60 may determine whether a macroblock (target block) to be encoded is located in an upper region of the screen based on the critical region set by the prediction mode determiner 80. When the target block is located in the upper area of the screen, as in step S111, the H.264 encoder 60 may predict the target block in a 16x16 mode.

If the target block is not located in the upper area of the screen, in step S103, the H.264 encoder 60 may determine whether the target block is located in the lower area of the screen. When the target block is located in the lower area of the screen, the H.264 encoder 60 may perform step S111.

When the target block is not located in the lower area of the screen, the H.264 encoder 60 determines that the target block is located in the middle area, and determines the prediction mode based on the threshold value set through the prediction mode determination unit 80. have. In step S105, the H.264 encoder 60 may designate reference blocks located adjacent to the target block, and calculate a variance value for reference pixels that are in contact with the reference blocks. In step S107, the H.264 encoder 60 may determine whether the calculated variance value is greater than a threshold value set by the prediction mode determiner 80. When the calculated variance value is less than the threshold value, the H.264 encoder 60 may predict the target block in a 16x16 mode as in step S111, and if the calculated variance value is greater than the threshold value, the H.264 encoder 60 As in step S109, the target block may be predicted in a 4x4 mode.

8 is a diagram for explaining the effect of the image processing method of the present invention.

Referring to FIG. 8, a screen 801 is a screen displaying experimental result values measured by encoding an arbitrary image according to the image processing method of the present invention. The above-described experimental result values include a result of encoding an image obtained using a 384x288 far-infrared ray sensor through an existing method conforming to the H.264 standard and a method to which the image processing method of the present invention is applied.

Referring to the table shown on screen 801, the result value corresponding to Situation 1 is the experimental result value measured for the image obtained in the situation where the vehicle 100 is traveling on an arbitrary straight road. The result of the experiment was measured on the image obtained in the driving situation, the situation 3 is driving the curved road, and the situation 4 is the driving situation in an arbitrary apartment complex. Screen 803 is a screen showing the image obtained in situation 2. In the case of using the encoding method to which the image processing method of the present invention is applied on average based on the experimental result values, the encoding efficiency is relatively reduced by about 0.6% compared to the conventional encoding method conforming to the H.264 standard, but encoding It can be seen that the speed is improved by more than about 20%.

As described above, the encoding method to which the image processing method of the present invention is applied reflects that an image to be encoded is a far-infrared image in front of a vehicle, so that a prediction mode to be applied in the encoding process can be determined early. As the prediction mode to be applied to the encoding process is determined early, the encoding speed may be improved. In addition, by determining a prediction mode in consideration of the characteristics of an image, it is possible to minimize a decrease in encoding efficiency. In addition, the encoding method to which the image processing method of the present invention is applied conforms to the encoding standard (ISO/IEC 14496-10), so that conventional decoders can operate normally.

In the present specification and drawings, a preferred embodiment of the present invention has been disclosed, and although specific terms are used, these are merely used in a general meaning to easily explain the technical content of the present invention and to aid understanding of the present invention. It is not intended to limit the scope of. In addition to the above-described embodiments, it is obvious to those of ordinary skill in the art that other modifications based on the technical idea of the present invention can be implemented.

100: vehicle 110: far infrared sensor
160: control unit

Claims (12)

In the vehicle,
A far-infrared sensor for collecting a front image of the vehicle; And
A threshold region and a threshold value for determining the complexity of a screen are set based on encoding information of the first frame constituting the image, and the critical region and the threshold value when the second frame constituting the image is predictively encoded A vehicle including a; a control unit controlling to determine a size of a target block to be designated as a coding unit by using a threshold value. The method according to claim 1,
The control unit
A vehicle that designates the first frame a certain number of times at a certain time interval, and controls to designate a frame temporally located behind the first frame as the second frame. The method according to claim 1,
The encoding information of the first frame is
A vehicle including sizes of blocks included in the first frame designated as the coding unit and variance values of pixels included in the blocks with respect to reference pixels to which a reference block adjacent to the blocks abuts. The method of claim 3,
The control unit
The critical area is set as an area corresponding to a boundary between first blocks having a relatively large size of the blocks and second blocks having a relatively small size of the blocks,
Designating a screen area in which the first blocks are located above the critical area as an upper area,
Designating a screen area in which the first blocks are located below the critical area as a bottom area,
A vehicle that designates a screen area excluding the upper area and the lower area as an intermediate area. The method of claim 4,
The control unit
When the number of the first blocks having the variance value as a specific value and the number of the second blocks are the same, the vehicle sets the specific value as the threshold value. The method of claim 5,
The control unit
The size of the target block located in the upper region and the lower region is set to be relatively large,
When the variance value of the target block located in the middle area is less than the threshold value, the size of the target block is set relatively large,
When the variance value of the target block located in the middle area is greater than the threshold value, the size of the target block is set relatively small. In the vehicle image processing method,
Collecting a front image of the vehicle;
Setting a threshold region and a threshold value for determining the complexity of a screen based on encoding information of a first frame constituting the image; And
And controlling a second frame constituting the image to determine a size of a target block to be designated as a coding unit using the threshold region and the threshold value during intra prediction encoding. The method of claim 7,
The step of setting the threshold area and the threshold value
Designating the first frame a predetermined number of times at a predetermined time interval; And
And controlling a frame located temporally behind the first frame to be designated as the second frame. The method of claim 7,
The step of setting the threshold area and the threshold value
The size of blocks included in the first frame designated as the coding unit and a variance value of pixels included in the blocks with respect to reference pixels contacting reference blocks adjacent to the blocks are determined in the first frame Including the encoding information of the; image processing method comprising a. The method of claim 9,
The step of setting the threshold area and the threshold value
Setting an area corresponding to a boundary between first blocks having a relatively large size of the blocks and second blocks having a relatively small size of the blocks as the critical area;
Designating a screen area in which the first blocks are located above the critical area as an upper area;
Designating a screen area in which the first blocks are located below the critical area as a bottom area; And
Designating a screen area excluding the upper area and the lower area as an intermediate area. The method of claim 10,
The step of setting the threshold area and the threshold value
And setting the specific value as the threshold value when the number of the first blocks having the variance value as a specific value and the number of the second blocks are the same. The method of claim 11,
Controlling to determine the size of the target block
Setting the size of the target block located in the upper region and the lower region to be relatively large;
Setting a size of the target block to be relatively large when the variance value of the target block located in the middle area is less than the threshold value; And
And setting a size of the target block to be relatively small when the variance value of the target block located in the middle region is greater than the threshold value.

Images