JPH07320056A - Vector image encoding device - Google Patents

Abstract

PURPOSE:To provide a vector image encoding device which is more efficient for an encoding system when vector data obtained by handwriting input are transmitted and stored, specially, for an encoding system which compresses the data by using a difference encoding method. CONSTITUTION:A numerical value to be encoded is inputted from an input part 401. A difference arithmetic part 402 finds the difference from a last input value. A vector size prediction part 403 predicts the size of subsequent vectors from a past input sequence and a code table selection part 40 selects a proper code table according to the predicted size of the vectors. A block relative coordinate arithmetic part 405 transforms vector data into a block number, a quadrant number, and relative coordinates. A block difference arithmetic part 406 finds the differences from a last block number and a last quadrant number. A difference encoding part 407 selects a code word according to the difference between the block numbers and the difference between quadrant numbers. A code word selection part 408 selects which of the code word and inputted numerical value itself is outputted to an output part 409.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image coding apparatus for transmitting handwriting input by a tablet or the like via a communication line, or for recording image data composed of an input coordinate series on a disc. The present invention relates to a vector image coding device used for a device or the like.

[0002]

2. Description of the Related Art Generally, when transmitting image information composed of a handwritten input coordinate sequence by a tablet or recording an image, the amount of image information is enormous, whereas In view of the line speed and cost of the line used for the above, the capacity and cost of the recording medium, and the like, it is necessary to compress and encode the image information for transmission and recording.

In such a case, as a coding apparatus for image information having a high compression rate, the difference between successive vectors when an input coordinate series is treated as a vector is used, and generally in the case of input such as handwriting, between consecutive input vectors. The difference vector value obtained by taking the difference has a smaller absolute value than the original vector, and it has the property that the direction does not change so much between consecutive vectors. It can be carried out. As an example of an encoding method using such a method, for example, CCI
TT Recommendation T. 150, JT-T150 in Japan
A coding method according to the standard is known.

A conventional method will be described with reference to FIG.
Input from the input unit 101, such as a tablet of a handwriting input terminal, is input as continuous coordinate data. The continuous coordinate data is converted into vector data in the difference calculation unit 102 by a calculation such as the following equation.

D _i = P _i -P _{i -1} where P _i is the i-th coordinate data, D _i is the i-th vector data, and i is an integer. The obtained vector data is sent to the block / relative coordinate calculation unit 103,
It is converted into a representation of a set of block number, quadrant number, and relative coordinate.

FIG. 2 is a diagram for explaining the conversion of vector data into block numbers, quadrant numbers and relative coordinates.
First, the vector data includes the first to the first as shown in FIG.
It is determined which one of the fourth quadrants is included. After the quadrant is determined, regarding the absolute value of each of the X and Y components of the vector data, which vector among the blocks is divided into the space to which the absolute value of the vector belongs increases as shown in FIG. 2B. Find which block contains the tip of. Further, the coordinate value of the tip of the vector in the block including the tip of the vector is determined with the point having the smallest X and Y coordinates in the block, that is, the point at the lower left of the block as the origin, and set as the relative coordinate. For example, the vector (7, 3) is represented by quadrant number 1, block number 5, and relative coordinates (3, 3).

The obtained quadrant number, block number, and relative coordinates are sent to the block difference calculation unit 104. The block difference calculation unit 104 obtains the difference between the block number and the quadrant number by the formula shown below.

BD_i= B_i-B_i-1 QD_i= Q_i-Q_i-1  BD here_iIs the difference of the i-th block number, B_iIs i
Th block number, QD_iIs the difference of the i-th quadrant number
Minutes, Q_iIs the i-th quadrant number, and i is an integer.
Obtained block number difference, quadrant number difference and relative
The coordinates are sent to the difference encoding unit 105.

The difference encoding unit 105 obtains a code word for a combination of the block number difference and the quadrant number difference from the code word table as shown in FIG. 3, and the code word selecting unit 106.
Output to. If the combination is not included in the codeword table, a special value indicating that the combination is not included in the codeword table is output to the codeword selection unit 106.

In the code word selecting unit 106, when the code word is present in the differential encoding unit 105, the code word and relative coordinates are output to the output unit 107, and when the code word is not present, the code word is output. Instead, a code sequence indicating that there is no code word is added and the coordinate value in the input unit 101 is output to the output unit 107 as it is.

[0011]

In the case of input from a handwriting tablet which is the target of differential encoding, the speed of input varies depending on the individual, so that the average between coordinate points input from the handwriting tablet is different. The distance is also different. Further, when the type of the handwriting input device used for input and the input coordinate values are used for communication and storage, the required accuracy varies depending on the final use.

In the above-mentioned conventional example, a fixed set of code words is used for encoding regardless of the sampling rate of the input device used for inputting coordinate values, the input speed, and the required output accuracy. However, there is a problem in that it is not possible to increase the transmission efficiency by selecting a set of code words that enables higher compression depending on the accuracy of the input point.

Further, when the display accuracy of the input device is different from that of the display device, there is a problem in that information for displaying in the same size as when input is not included.

Therefore, in the present invention, the code table is changed and encoded by the input data so that the compression rate is higher than that of the conventional encoding method with the simple structure, and the encoding is performed on the decoding side. It is an object of the present invention to provide a vector image encoding device that can perform appropriate decoding based on information contained in data and information about an output device.

[0015]

In order to achieve the above object, the present invention provides a vector image coding apparatus which treats an input coordinate sequence as a continuous vector in an image, and calculates the average of the input vectors. Vector size predicting means for predicting the size, code table determining means for determining a table to be used for encoding based on the average size of the vector predicted by the vector size predicting means, and the code table determining means. And a coding unit that performs coding based on the above table.

[0016]

The vector size predicting means accumulates past input sequences and predicts an average distance of vectors generated thereafter from the past input sequences. The code table determining means receives the output of the vector size predicting means and determines a table to be used as a code table from the table determined according to the vector size.

[0017]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 4 is a block diagram of a coding unit of an embodiment in which the vector image coding apparatus of the present invention is applied to a telewriting terminal device. The encoding unit includes an input unit 401, a difference calculation unit 402, a vector size prediction unit 403, a code table selection unit 404, a block / relative coordinate calculation unit 405, a block difference calculation unit 406, and difference encoding. Unit 407, codeword selection unit 408, and output unit 4
And 09.

In this encoding section, the handwritten input coordinates input from the input section 401 are converted into a vector coordinate sequence by taking the difference from the immediately preceding coordinates in the difference calculation section 402. The vector coordinate sequence is sent to the vector size prediction unit 403 and used for subsequent prediction of vector size.

The vector size prediction unit 403 stores the past input vector series, and performs an appropriate interpolation operation, for example, linear interpolation, from the past value and the current input to determine the size of the vector to be subsequently input. And the size of the vector to be predicted is output to the code table selection unit 404.

The code table selecting unit 404 uses the prediction of the vector size predicting unit 403 to determine a plurality of tables that are fixedly fixed in advance, or that the transmitting side sends to the receiving side in advance at the beginning of communication. The code table is determined by selecting from a plurality of tables in accordance with the standard as shown in FIG. 5, for example.

In FIG. 5, the table No. Is the number of the table used for determining the block number from the absolute value of the vector, and determines the minimum block size N when the block size that was fixed in the conventional example as shown in FIG. Then, the code table used to convert the vector absolute value created in accordance with the block number is selected. In the example of FIG. 5, when the average vector is large, it is set to select a table in which the range of the area having a small block number finally converted to a short codeword length is widened.

If the selected table is different from the currently used table, the block / relative coordinate calculation unit 405 is notified of the new table, and the output unit 409 outputs a code indicating the change of the code table. After the output, the number of the selected table is output.

The block / relative coordinate calculation unit 405 uses the code table selected by the codeword selection unit 404 to convert the vector data sent from the difference calculation unit 402 to the block number, quadrant number, and relative within the block. It is converted into coordinate values and output to the block difference calculation unit 406.

In the block difference calculation unit 406, the immediately preceding block number, the quadrant number and the newly obtained block number,
The block number difference and the quadrant number difference are obtained using the quadrant numbers. The obtained block number is sent to the differential encoding unit 407.

The differential encoding unit 407 selects one codeword for the set of the block number difference and the quadrant number difference, and outputs it to the code selection unit 408. When a codeword is defined by a combination of the difference value of the corresponding block number and the quadrant number difference, the codeword is transmitted to the codeword selecting unit 408, and the relative coordinate value is subsequently transmitted. To do. When the corresponding codeword is not defined, a special code string indicating that the codeword does not exist is sent to the codeword selecting unit 408.

In the code word selection unit 408, the differential encoding unit 4
If the output of 07 consists of codewords and relative coordinates,
The data is sent to the output unit 409 as it is. When the output of the differential encoding unit 408 is a code string indicating that no codeword exists, a code string indicating that absolute coordinates are to be transmitted is transmitted to the output unit 409, and the absolute value obtained from the input unit 401 is obtained. Coordinates are subsequently sent.

FIG. 7 is a block diagram of an embodiment of a decoding unit of a telewriting terminal device based on the vector image coding device of the present invention. The decoding unit is the input unit 7
01, a code determination unit 702, a code extraction unit 703, a quadrant / block calculation unit 704, and a vector coordinate calculation unit 70.
5, a code table selection unit 706, and an absolute coordinate calculation unit 707.
And an output unit 708.

In this decoding section, the code string inputted from the input section 701 is judged in the code table judging section 702 as to the code table used for coding. In the code table determination unit 702, when a code string indicating a change in the code table appears in the code string from the input unit 701, the code table number used for the subsequent coding is obtained from the code string that follows. The obtained code table number is sent to the code table selecting unit 705. Code strings other than the code table changes are sent to the code extraction unit 703 without any changes.

The code table selecting unit 705 includes a code determining unit 70.
An appropriate block size is determined based on the code table number sent from the terminal 2 and the displayable accuracy of the terminal itself, and the code table is selected.

The code extraction unit 703 reads the input code sequence in order, and first determines from the first code sequence of the data whether the data is based on absolute coordinates or a code word based on differential encoding. When the data is based on absolute coordinates, the absolute coordinates are calculated and the obtained result is sent to the output unit 708. In the case of a code word by differential encoding, a quadrant number difference, a block number difference,
And the relative coordinates are calculated and sent to the quadrant / block calculation unit 704.

In the quadrant / block calculation unit 704,
A new quadrant number and a block number are obtained by adding the difference in the quadrant number and the difference in the block number calculated in the code extracting unit 703 to the immediately preceding quadrant number and block number, respectively, and the vector coordinate calculating unit 706 together with the relative coordinates.
Send to.

The vector coordinate calculation unit 706 converts the values sent from the quadrant / block calculation unit 704 into vector coordinates in the reference coordinate system using the code table selected by the code table selection unit 705.

The absolute coordinate calculation unit 707 obtains a new coordinate value by adding the vector obtained by the vector coordinate calculation unit 706 to the immediately preceding coordinate value and outputs it to the output unit 710.

[0035]

In the vector image coding apparatus according to the present invention, the code used for coding is switched according to the change amount of the input vector data, so that the code is more flexible than the coding by the fixed code. It is possible to select a set of code words that can obtain a high compression rate during conversion.

Further, by applying the vector image coding apparatus utilizing the present invention to a handwriting input data storage device using a tablet or the like as an input means and a handwriting input data communication device, an appropriate compression according to the capability of the terminal is performed. It is possible to perform efficient storage and communication.

[Brief description of drawings]

FIG. 1 is an explanatory diagram of a configuration example of a conventional vector image encoding device.

FIG. 2 is an explanatory diagram of block numbers / quadrant numbers.

FIG. 3 is an explanatory diagram of an example of a code table used for encoding.

FIG. 4 is a block configuration diagram of an encoding device according to an embodiment of the present invention.

FIG. 5 is a table for determining a block size from the size of a vector.

6 is an explanatory diagram of a minimum block size in FIG.

FIG. 7 is a block configuration diagram of a decoding device according to an embodiment of the present invention.

[Explanation of symbols]

 Reference numeral 101 input unit 102 difference calculation unit 103 block / relative coordinate calculation unit 104 block difference calculation unit 105 difference encoding unit 106 codeword selection unit 107 output unit 401 input unit 402 difference calculation unit 403 vector size prediction unit 404 code table selection unit 405 Block / relative coordinate calculation unit 406 Block difference calculation unit 407 Difference encoding unit 408 Code word selection unit 409 Output unit 701 Input unit 702 Code determination unit 703 Code extraction unit 704 Quadrant / block calculation unit 705 Vector coordinate calculation unit 706 Code table selection Part 707 Absolute coordinate calculation part 708 Output part

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification number Office reference number FI technical display location H04N 5/92 7/24 H04N 7/13 Z

Claims (3)

[Claims] 1. A vector image encoding apparatus for treating an input coordinate sequence as a continuous vector sequence in an image, a vector size prediction unit for predicting an average size of an input vector, and the vector size prediction. Code table determining means for determining a table to be used for encoding based on the average magnitude of the vector predicted by the means, and encoding means for performing encoding based on the table determined by the code table determining means. A vector image encoding device characterized by the above. 2. The vector size predicting means stores a past input vector series, and predicts the size of a vector to be subsequently input by performing an appropriate interpolation operation from past values and the present input. The vector image coding apparatus according to claim 1, wherein 3. A vector image encoding apparatus for treating an input coordinate sequence as a continuous vector sequence in an image, an input unit for inputting a numerical value to be encoded, an input value from the input unit and a previous input. Difference calculation unit that takes the difference from the value and converts it to a vector coordinate sequence, vector size prediction unit that predicts the size of the vector from the past vector coordinate input sequence, and an appropriate code table from the predicted vector size And a vector coordinate sequence sent from the difference calculation unit to a block number, a quadrant number, and a relative coordinate value within the block, using the code table selected by the codeword selection unit. Block / relative coordinate calculation unit, the immediately preceding block number, the quadrant number and the newly obtained block number, the block number difference using the quadrant number, and the quadrant number A block difference calculation unit for obtaining the difference between the block number difference and the quadrant number difference for selecting a code word, the selected code word, and outputs any of the input numerical value itself A vector image encoding device, comprising: a code selecting unit that selects whether to perform.