JP2642545B2 - How to convert visual characteristics into block orthogonal transform domain - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is used in an image compression encoding apparatus, and particularly when performing quantization or evaluation in consideration of human visual characteristics, visual characteristics (frequency characteristics).
To a block orthogonal transform domain.

[0002]

2. Description of the Related Art As a conventional visual characteristic area conversion method, for example, a document: "Optimal quantization of DCT coefficients in consideration of visual characteristics in variable length coding" (IEICE Technical Report IE90-101) Was disclosed. FIG. 3 is a diagram for explaining this conventional method. Hereinafter, the processing procedure of the conventional method will be described. In FIG. 3, in order to obtain visual characteristics of an orthogonal transform area having an N × N block size, first, block data of an orthogonal transform area in which all N × N block values are A is input.

[0003] The block data is orthogonally transformed to obtain a spatial domain impulse. At this time, the coefficient A is adjusted so that the height of the impulse becomes 1.

[0004] The impulse in the spatial domain is transformed to the spatial frequency domain using a discrete Fourier transform (DFT).

The spatial frequency domain data is also multiplied by the spatial frequency domain visual characteristic H (k, m).

[0006] This data is subjected to an inverse discrete Fourier transform (I
(DFT) to obtain a spatial impulse response based on visual characteristics.

[0007] Finally, the spatial impulse response is transformed into a predetermined orthogonal transform region by performing an inverse orthogonal transform,
Obtain visual characteristics in the orthogonal transform domain. In this case, since the visual characteristics are represented by complex numbers, the norm thereof is obtained and further normalized by the coefficient A to obtain the normalized visual characteristics in the orthogonal transform domain.

Table 1 shows an example of visual characteristics in an 8 × 8 block when the visual distance L is L = 1024 (pixels) and the discrete cosine transform (DCT) is used as the orthogonal transform. is there.

[0009]

[Table 1]

[0010]

However, when the conventional transform method is applied to the block orthogonal transform domain, the spatial impulse response due to the visual characteristic which should originally spread over many blocks is folded back into one block, and an accurate impulse response is obtained. There has been a problem that a response cannot be obtained, and thus accurate conversion of visual characteristics cannot be performed.

According to the present invention, since the spatial impulse response due to the visual characteristics described above is folded back into one block, an accurate impulse response cannot be obtained, and therefore, the conversion of the visual characteristics cannot be accurately performed. It is an object of the present invention to provide a conversion method that removes adverse effects and accurately converts visual characteristics.

[0012]

According to the present invention, in order to solve the above-mentioned problems, (a) the M × N spatial impulse in the spatial region is extended to mM × nN, and (m, n, M, N : A positive integer. (B) The spatial impulse expanded in the area is converted into a spatial frequency domain, and (c) the spatial impulse converted into the spatial frequency domain is multiplied by a visual characteristic in the spatial frequency domain, and is inversed in the spatial domain. (D) transform the spatial impulse response into a block orthogonal transform domain, find the norm value of the same component of each block, and (e) convert the spatial impulse into a block orthogonal transform domain. (F) determining a visual characteristic in a block orthogonal transform domain based on a ratio of a norm value of the spatial impulse response to an absolute value of the spatial impulse. To.

[0013]

According to the present invention, when a spatial region having a spatial impulse having a block size of M × N is extended to a region of m × M × n × N by a variable magnification m, n which can be set according to the block size. In addition, the expansion magnification is selected so that no aliasing occurs in the spatial impulse response, and the peak point of the spatial impulse is located near the center of the space in the extended area, and the spatial impulse response is converted to the spatial frequency domain in the extended area. Later, the visual characteristic is given, and the ratio between the norm value of the spatial impulse subjected to the block orthogonal transform and the absolute value of the spatial impulse subjected to the block orthogonal transform is obtained. This enables accurate conversion of visual characteristics without distortion caused by folding back.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a functional block diagram of an apparatus showing an embodiment of the present invention. In FIG.
Block size (M and N are M, N for convenience of high-speed conversion, etc.)
= 2P is often set). The peak point of the impulse is located near the center of the block in consideration of the characteristics of the impulse response described later.

The spatial area expanding means 1 expands the spatial area into the area of mM × nN centering on the block of M × N of the input 101. However, m and n are positive integers, and a suitable value according to the block size is selected as described later so that no aliasing occurs in the impulse response.

The discrete Fourier transform (DFT) means 2 converts the extended mM × nN spatial domain 102 into a spatial frequency domain.

The visual characteristic adding means 3 adds the visual characteristic table 9 to the spatial impulse 103 converted into the spatial frequency domain.
Characteristics 111 in the spatial frequency domain output from
Multiply by

The inverse discrete Fourier transform (IDFT) 4 converts the spatial impulse 104, which has been added to the visual characteristics and is converted into the spatial frequency domain, into the spatial domain again. As a result, a spatial impulse response 105 based on visual characteristics is obtained.

The block orthogonal transform means 5 converts the spatial impulse response 105 into an M × N block orthogonal transform area, and also performs block orthogonal transform of the spatial impulse input 101 for only one block.

The norm calculation means 6 performs a norm calculation of the same component 106 of each block subjected to the block orthogonal transformation, and the absolute value means 7 performs an absolute value calculation of the orthogonally transformed space impulse 108.

The comparison normalizing means 8 calculates the ratio between the orthogonal transform value 107 of the spatial impulse to which the visual characteristic is added and the orthogonal transform value 109 of the spatial impulse, normalizes the ratio as necessary, and performs visualization in the block orthogonal transform region. 110 is output.

Next, the details and procedure of the method of converting the visual characteristics into the block orthogonal transform area will be described with reference to FIG.

First, a spatial impulse [X (k, l)] 31 having a block size of M × N is input. However, [・]
Represents a matrix, and k and l take discrete values of 0 ≦ k <M and 0 ≦ l <N.

[0025]

(Equation 1)

The spatial impulse [X (k, l)]
Is expanded to (m × n) times (where m, n ≧ 1) such that is the center, and all data in the expanded area is set to “0”. The expansion magnifications m and n are, for example, m, n ≧ 3 when M, N = 8, and m, n ≧ 2, M, N ≧ when M, N = 16.
In the case of 32, if m and n ≧ 1, the spatial impulse response is 1
Good results are obtained without folding into one block.

The spatial impulse 32 whose area has been extended is subjected to a discrete Fourier transform (DFT). This DFT is represented by the following equation (2).

[0028]

(Equation 2)

[F (u, v)] 33 shows the D of the visual characteristic.
The value 35 is multiplied by an FT value [HF (u, v)] 34, and the value 35 is subjected to inverse discrete Fourier transform (IDFT) to return to the space domain. This I
DFT is represented by the following equation (3). The obtained [X _H (k,
l)] is the response due to the visual characteristics of the spatial impulse.

[0030]

(Equation 3)  

The impulse response [X _H (k,
1)], and the spatial impulse [X (k, l)] is similarly subjected to block orthogonal transformation. This block orthogonal transform is represented by the following equations (4) and (5), respectively. Here, (i, j) in Expression (4) represents a block number, [C] is an arbitrary orthogonal transformation function matrix, and [C ^t ] is its transposed matrix.

[CH ^{(i, j)}(u, v)] = [C^t] ・ [X_H(i · M + k, j · N + l)] · [C] (4) [C_X(u, v)] = [C^t] ・ [X (k, l)] ・ [C] (5) [C_H ^{(i, j)}(u, v)] 37 is [C_X(u, v)]
Showing visual characteristics, but to image compression and evaluation of visual characteristics
Considering the application of [C_H ^{(i, j)}(u, v)]
Norm of one component ‖C_H(u, v) 対 象 by the following equation (6)
Defined as the square root of the sum of squares of the absolute values of the elements of the matrix
Justice, [‖C_H(u, v) ‖] 38 and [C
_X(u, v)], which is the absolute value of each element of [| C_X(u, v)
|] Is obtained.

[0033]

(Equation 4)

Finally, [‖C given by equation (7)
The ratio between each element of [ _H (u, v)}] and [| C _X (u, v) |] is obtained, and is set as a visual characteristic in the block orthogonal transform domain.
The visual characteristic [H _C ] 41 can be normalized or multiplied by K as necessary. H _C (u, v) = { _{C H} (u, v)} / | C _X (u, v) | (7) L = 1024, S = 3, M, N = 8, m, n = 3 Table 2 shows an example of converting the visual characteristics in the case into a block cosine transform region.
To Table 5 below. Table 2 shows the spatial impulse response [X _H ], and Table 3 shows [C _H ] obtained by subjecting [X _H ] to block cosine transform. Table 4 shows [C _x ] obtained by subjecting a spatial impulse to block cosine transform, and Table 5 shows visual characteristics [H _C ] in the block cosine transform domain.

[0035]

[Table 2]

[0036]

[Table 3]

[0037]

[Table 4]

[0038]

[Table 5]

[0039]

As described above, according to the present invention, the spatial impulse response of the visual characteristic is accurately represented, and the influence of the spread of the spatial impulse response on the surrounding blocks is appropriately represented. It is possible to realize the conversion of the visual characteristics into the block orthogonal transform area.

Also, in an image compression coding apparatus or the like, since a quantization method or a coding method suitable for the visual characteristics of an image plays an important role, the compression encoding efficiency of an image is greatly increased by using accurate visual characteristics. And the quality of the image can be improved.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of an apparatus showing one embodiment of the present invention.

FIG. 2 is an explanatory diagram of a conversion method according to the present invention.

FIG. 3 is an explanatory diagram of a conventional system.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Spatial area expansion means 2 DFT means 3 Visual characteristic addition means 4 IDFT means 5 Block orthogonal transformation means 6 Norm calculation means 7 Absolute value calculation means 8 Comparison / normalization means 9 Visual characteristic table

Claims (1)

(57) [Claims] (A) expanding an M × N spatial impulse in a spatial area to mM × nN (m, n, M, N: a positive integer); and (b) expanding the spatial impulse in the spatial area. Is converted to the spatial frequency domain, and (c) the spatial impulse converted to the spatial frequency domain is multiplied by a visual characteristic in the spatial frequency domain, and inversely converted to the spatial domain to obtain a spatial impulse response. (E) converting the spatial impulse to a block orthogonal transform domain to determine the absolute value of each component, and (f) calculating the absolute value of each component. A method for converting visual characteristics into a block orthogonal transform region, wherein a visual characteristic in a block orthogonal transform region is obtained based on a ratio between a norm value of a response and an absolute value of the spatial impulse.