JP2994643B2 - Image transmission apparatus and method - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image transmission apparatus for transmitting image data and a method thereof. [Prior Art] Conventionally, for example, a time axis conversion band compression method (hereinafter, referred to as TAT) has been announced as an information compression transmission method, but this TAT method is used when information signal band compression is performed. This is a method of compressing and transmitting an information signal by utilizing the fact that the density of the signal varies depending on the location. FIG. 2 shows the principle in the case of using the TAT method in one-dimensional signal processing. In FIG. 2, first, the original signal is divided for each predetermined amount of information as shown by a dotted line, and it is determined whether the information is coarse or dense for each divided group. In the group determined to be dense, all of the data obtained by sampling the original signal is transmitted as transmission data, and in the group determined to be coarse, only part of all data is determined to be transmission data. , And others are not transmitted as thinned data. Note that the data shown with a circle in the figure is data to be transmitted (transmission data), and the data shown with a cross is data that is not transmitted (thinned-out data). By transmitting the transmission data indicated by the mark at regular intervals, the number of data transmitted per unit time is reduced, and the band of the transmission information signal is compressed. In the data transmitted as described above, the thinned-out data not transmitted at the time of decoding is approximately restored using the transmitted transmission data to obtain interpolation data (indicated by a circle in the drawing). In addition, this interpolation data corresponds to a part where information is coarse,
Since the data is restored as data that is very similar to the thinned data, the substantial information amount does not change as compared with the case where all the data is transmitted, and the transmission band of the information signal is largely compressed. At this time, whether to transmit all data or a part of the data in each group is determined by examining the details of the original signal, and this determination information is transmitted simultaneously as a transmission mode information signal. In the case of an image information signal, the transmission band of the image information can be compressed by changing the sampling interval in the vertical direction as well as the sampling interval in the horizontal direction and performing two-dimensional processing. When processing a signal two-dimensionally, such as an image information signal, one screen is divided into m × n groups of pixels, and the density of the image is determined for each group. And
In the group determined to be dense, data obtained from all pixels is transmitted as transmission data, and in the group determined to be coarse, data obtained from some of the pixels is transmitted as transmission data, The data of the remaining pixels is not transmitted as thinned data. Here, the case where data obtained from all pixels is transmitted is referred to as E mode, and the case where only data for some pixels is transmitted is referred to as C mode. Pixels to which data is transmitted in each transmission mode are not transmitted. The relationship with the pixels is as shown in FIG. FIGS. 3A and 3B show pixels to which E-mode and C-mode data are transmitted in the 4.times.4 pixel component loop, where FIG. 3A shows the E mode and FIG. One screen of the image information to be transmitted is divided into 4 × 4 pixel groups in order from the upper left to the lower right of the screen, and the two types of transmission modes as described above are set according to the density of the image for each group. Select and transmit according to the selected transmission mode. FIG. 4 shows that the NTSC television screen is subjected to component demodulation according to the above-described method, and one field of the time-division multiplexed television screen is divided into 4 × 4 pixel groups. FIG. 4 is a diagram in which transmission modes of an E mode and a C mode shown in FIG. 3 are assigned. In addition, ○
The mark is a pixel to which data is transmitted, and the cross is a pixel to which data is not transmitted. By transmitting pixel data transmitted in this manner at regular intervals, the transmission band is compressed and transmitted. In addition, since the data of the pixel which is not transmitted is approximately restored by using the data of the neighboring transmitted pixel at the time of decoding, the substantial information amount does not change compared with the case where all the data is transmitted, and the information is not changed. This means that the signal transmission band has been greatly compressed. Further, when information is transmitted using the information compression transmission method as described above, particularly when information is recorded on a recording medium or the like and reproduced, information in a different transmission mode is recorded on the recording medium. Therefore, if the mode information is erroneously detected, the original signal cannot be restored, but in order to solve this problem, the basic pixel group and the additional pixel group are separated and transmitted by a predetermined amount. A method has been proposed. Here, the basic pixel indicates a pixel transmitted in both the E mode and the C mode, and the additional pixel indicates a pixel transmitted only in the E mode. According to this method, even when the mode information is erroneously detected, only the information signal of the separated and transmitted basic pixel group is taken out, and the information signal can be restored by interpolating the additional pixels. [Problem to be Solved by the Invention] In the conventional example, to obtain a higher compression ratio, E
It is necessary to change the distribution ratio between the mode and the C mode in a direction in which the number of additional pixels decreases, that is, in a direction in which the number of C modes increases. However, there is a problem in that the less information transmitted in the E mode, the more the image information signal is degraded in image quality. An object of the present invention is to provide an image transmission apparatus and method capable of compressing and transmitting image data without increasing the qualitative deterioration of the image data in the background described above. [Means for Solving the Problems] An image transmission apparatus according to the present invention comprises: an input unit for inputting image data; a dividing unit for dividing the image data input by the input unit into blocks each including a plurality of pixels; The image data of a block represents basic information in the block, represents first coded data having a fixed number of bits regardless of the block, and represents information other than the basic information in the block, and corresponds to the block. Encoding means for encoding the encoded data into second encoded data having a variable number of bits, and the first encoded data encoded by the encoding means is encoded before the second encoded data. And transmission means for arranging and transmitting the data. Further, the image transmission method according to the present invention, the image data is input, the input image data is divided into blocks consisting of a plurality of pixels, the image data of each block represents the basic information in the block, The first encoded data having a fixed number of bits regardless of the block and the second encoded data representing information other than the basic information in the block and having a variable number of bits according to the block are encoded. Wherein the encoded first encoded data is arranged and transmitted before the second encoded data. Examples Hereinafter, the present invention will be described based on examples. Here, as an embodiment of the present invention, a case where the present invention is applied to a VTR will be described as an example. The information signal handled here is a signal obtained by subjecting the NTSC television signal to component demodulation and time division multiplexing. When a 1-field television screen is divided into a plurality of pixel groups, the group is divided into 4 × 4 pixels. It shall consist of groups.
The pixel group is composed of a sub-pixel group including one basic pixel and three additional pixels as shown in FIG. FIG. 1 is a diagram showing a schematic configuration when the present invention is applied to a VTR recording system. In the embodiment shown in FIG. 1, when the transmission is performed in the E mode, the basic pixel data is directly encoded into a predetermined number of quantization bits (for example, 8 bits) and transmitted. The data is transmitted as data obtained by performing differential encoding for performing variable length encoding on the difference value from the basic pixel of the group. When transmission is performed in the C mode, the band-limited basic pixel data is transmitted in the same manner as the E mode basic pixel data. In FIG. 1, the input analog video signal for one field is converted to a digital analog (A / D) converter 1.
Is converted into a digital video signal, and supplied to the switch 2, the pre-filter 3, and the block distortion calculating circuit 18 as an E-mode digital video signal. E-mode digital input from the pre-filter 3
A two-dimensional low-pass filter that removes high-frequency components of the video signal and averages the digital video signal. The E-mode digital video signal averaged by the pre-filter 3 as shown in FIG. 3A is output by the thinning circuit 5 to the C-mode digital video signal of 1/4 pixel as shown in FIG. 3B. -A process of thinning out a video signal is performed. The C-mode digital video signal processed in the thinning circuit 5 is stored in the fourth memory 6.
And is supplied to a block distortion calculating circuit 18. When the switch 2 is connected to the G side in the figure, the digital video signal is supplied to a switch 9. When the switch 9 is connected to the I side in the figure, the input signal is supplied to the first memory 11, and when the switch 9 is connected to the H side in the figure, the sub-pixel group delayed by the delay circuit 15 is supplied. A difference signal between the digital video signal corresponding to the basic pixel and the digital video signal corresponding to the input additional pixel is output by the subtractor 27, is subjected to variable-length encoding in the encoder 16, and is supplied to the second memory 10. Is done. When the switch 2 is connected to the F side in the figure, the digital video signal is subtracted from the digital video signal corresponding to the basic pixel of the sub-pixel group held in the first memory 11 by the subtractor 28. The operation is performed, and the difference signal is output from the subtractor. And
The difference signal is variable-length coded by an encoder 17 and supplied to a third memory 8. Here, the encoders 16 and 17 output 4 bits when the input difference signal is, for example, near the threshold value for mode determination, and 8 bits otherwise.
The difference signal corresponding to an additional pixel that frequently appears in the E mode is converted into a short code. Thereby, the average code length of the difference signal can be shortened. The switching operation of the switches 2 and 9 is controlled by various synchronization signals generated by the synchronization signal generation circuit 7. From the synchronizing signal generating circuit 7, a horizontal synchronizing signal synchronized with the input timing of the input video signal for one line is input to the switch 2, and a pixel synchronizing signal synchronized with the input timing of the input video signal for one pixel is input to the switch 9. Is done. By the operation of switches 2 and 9, the E-mode digital video signal as shown in FIG. 3 (a) is divided into 4 × 4 pixel groups as shown by ◎, △ and に お い て in FIG. , The first memory 11 stores a signal corresponding to the pixel indicated by ◎, the second memory 10 stores a signal obtained by variable-length encoding the difference value between the pixel indicated by △ and the basic pixel, and the third memory 8 stores a signal obtained by performing variable-length encoding on the difference value between the pixel indicated by the triangle and the basic pixel. The E-mode and C-mode digital video signals are input to the block distortion calculating circuit 18. After the C-mode digital video signals are subjected to interpolation processing,
For each group, an image signal based on the E-mode digital video signal is compared with an image signal based on the interpolated C-mode digital video signal, and error information is calculated, and the error information is used as a block distortion value. Code length memory
19 and the distortion memory 20. The code length memory 19 sums up the code lengths of all the additional pixels of the pixel group having the block distortion value, and stores the code length information for each block distortion value. The mode setting circuit 4 generates a mode information signal for allocating a transmission mode to the group so that transmission is performed in the E mode when the block distortion value is larger than the set block distortion threshold value, and when the block distortion value is smaller than the set block distortion threshold value. I do. That is, this mode information signal is generated such that a group of dense portions on the screen of one field corresponds to a mode information signal of E mode, and a group of coarse portions corresponds to a mode information signal of C mode. Here, the threshold value is also set in the mode setting circuit 4 as follows. A code length information signal calculated for each block distortion value in units of one field is input from the code length memory 19, and the code length information signals are sequentially added in descending order of the block distortion value. A predetermined amount (for example,
The above-described addition operation of the code length information signal is repeated within a range not exceeding the size of the recording area of the signal corresponding to the additional pixel), and after the operation is completed, the threshold value is set using the block distortion value. As a result, a threshold value is set such that the input video signal is transmitted in the E mode as much as possible within the recordable range of the additional pixel. The mode information signal generated in the mode setting circuit 4 is stored in the mode memory 12. FIG. 7 shows an example of the arrangement of the mode information signal set on the screen as described above. The arrangement of the pixels to be transmitted according to the mode information signal arranged in this manner on the screen is as follows. As shown in FIG. However, what is actually transmitted as described above is ninth.
The information of the basic pixels is stored in the first memory 11 and the fourth memory 6 with the information of all the basic pixels in each group of the screen as shown in the figure and the information of the additional pixels in the group set to the E mode as shown in FIG. Since the difference information of the additional pixel is stored separately in the third memory 8 and the second memory 10, the information stored in these memories is stored in the eleventh memory according to the mode information.
The output is as shown in the figure. Hereinafter, the first memory 11, the second memory 10, the third memory 8,
The read operation of the fourth memory 6 will be described. Since the mode information generated by the mode determination circuit 4 is stored in the mode memory 12, the memory control circuit 13
Reads out the mode information signal stored in the mode memory 12. The read mode information signal is supplied to the memory control circuit 13, the address operation circuit 14, and the switch 21. The memory control circuit 13 supplies a basic pixel information signal in the E mode from the first memory 11 and a basic pixel information signal in the C mode from the fourth memory 6 to the switch 21 in response to the input mode information signal. The switch operation is performed in accordance with the information signal, and the basic pixel information signal as shown in FIG. The address operation circuit 14 operates the read addresses of the third memory 8 and the second memory 10 according to the input mode information signal, and controls the switch of data from the two memories. E in the information signal
When the mode is specified, the address of the pixel data to be output is calculated and output. In the C mode, the address is changed so that a predetermined amount of pixel data is skipped. The pixel data read by the arithmetic circuit 14 is supplied to the switch 22. The switch 22 is switched for each line in synchronism with the horizontal synchronizing signal generated from the synchronizing signal generating circuit 7. By this switching operation, the output of the switch 22 becomes an additional pixel information signal as shown in FIG. Are supplied to the switch 23. A switch 23 is switched every 0.5 field in synchronization with the synchronizing signal generated by the synchronizing signal generating circuit 7. The signal output by this switching operation is composed of basic pixel data and additional pixel data as shown in FIG. The data is separated and output in a time series, supplied to a digital / analog (D / A) converter 24, converted into an analog signal again, and band-limited through a low-pass filter (LPF) 25.
It is recorded in the recording section 26 together with the mode information signal. As described above, the image information signal is divided into the fixed-length coded information signal of the basic pixel and the variable-length coded difference information signal of the additional pixel of the group set in the E mode for all the divided groups. Since the transmission information signal is transmitted, it is possible to compress the transmission information signal while suppressing the quality deterioration, and it is possible to compose at least the information signal of the basic pixel even if the mode information signal is erroneous at the time of decoding. FIG. 12 is a diagram showing a schematic configuration when the present invention is applied to a VTR reproduction system. The operation during reproduction will be described below with reference to FIG. In FIG. 12, when the reproduction operation is selected by the system controller 42, the pixel information signal and the mode information signal reproduced by the reproduction section 29 are input to the A / D converter 30 and the memory control circuit 34, respectively. The pixel information signal is input to a switch 31 that has been converted from an analog signal to a digital signal by an A / D converter 30.
A switch 31 is supplied with a synchronization signal synchronized with a 1/2 field period generated in the reproducing unit 29. The digital pixel information signal recorded by the switch 31 as shown in FIG. And the information signal of the additional pixel.
33 and memorize it. The switching operation of the switch 31 is performed in synchronism with the synchronizing signal from the L side in FIG.
Switch to the side. On the other hand, the reproduced mode information signal is transmitted to the memory control circuit 34.
The memory control circuit 34 outputs the basic pixel memory 32 and the additional pixel memory 33 according to the input mode information signal.
, The stored pixel information signal is read out from each. The basic pixel information stored in the basic pixel memory 32 includes E-mode information and C-mode information. Therefore, the basic pixel information in the C mode is interpolated by the switch 35 using the transmitted pixel information in the C mode interpolation circuit 36, and the basic pixel information in the E mode is directly stored in the field memory 35. To memorize. That is, the reproduced mode information signal is input to the switch 35, and according to the mode information, to the C side in the figure in the case of the C mode, and to the E side in the figure in the case of the E mode. It is controlled to be connected. The E-mode basic pixel information read from the basic pixel memory 32 is also supplied to a delay circuit 38, and after being delayed by the delay circuit 38, is supplied to an adder 39. The other input of the adder 39 is supplied with the difference pixel information read from the additional pixel memory 33 and decoded from the variable length code to the fixed length code to the decoder 40. The addition is performed by the adder 39, and the additional pixel information in the E mode is restored and stored in the field memory 37. As described above, after one field of the pixel information signal is stored in the field memory 37, the pixel information signal stored in the field memory 37 is read out as an image information signal by the control signal from the memory control circuit 34 and is read. / A converter 41
Is output after being converted into an analog video signal. In the present embodiment, the difference value between the additional pixel data and the basic pixel of each sub-pixel group is variable-length coded, but the difference value between the additional pixel data is set to a variable-length code. Is also good. The variable-length encoding method may be, for example, run-length encoding or the like other than the above method, but is not limited thereto. In the present embodiment, a VTR has been described as an example. However, the present invention can be applied to any device that transmits an information signal, and can be applied to, for example, a disk device or a transmission device. [Effects of the Invention] As described above, according to the present invention, it is possible to prevent image deterioration without increasing the code amount, and to improve claims for reproducing images on the decoding side.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic structural view of a case where the present invention is applied to a VTR recording system as one embodiment of the present invention. FIG. 2 is a principle diagram when the TAT method is used in one-dimensional signal processing. 3A is a diagram showing a pixel pattern transmitted in the E mode in a 4 × 4 pixel group, and FIG. 3B is a diagram showing a pixel pattern transmitted in the C mode. FIG. 4 shows a one-field television screen of a time-division multiplexed signal obtained by component demodulation of an NTSC television signal.
FIG. 11 is a diagram illustrating a case where the image data is divided into four pixel groups and transmission modes of an E mode and a C mode are assigned to each group. FIG. 5 is a diagram showing 2 × 2 sub-pixel groups forming a 4 × 4 pixel group. FIG. 6 is a diagram showing the correspondence between 4 × 4 pixels in FIG. 1 and a memory for storing the pixel information. FIG. 7 is a diagram showing an example of the arrangement of the mode information signal on the screen. FIG. 8 is a diagram showing an arrangement on the screen of pixels to be transmitted according to the mode information signal arranged as shown in FIG. FIG. 9 is a diagram showing the arrangement of basic pixels in the pixels on the screen. FIG. 10 is a diagram showing the arrangement of additional pixels in the pixels on the screen. FIG. 11 is a diagram showing a transmission mode of information signals of a basic pixel and an additional pixel. FIG. 12 is a diagram showing a schematic configuration of the reproducing system of FIG. 2,9,21,22,23,31,35 ... Switch, 5 ... 3/4 thinning circuit, 6,8,10,11 ... Memory, 4 ... Mode judgment circuit, 12
…… Mode memory, 13,34 …… Memory control circuit, 15,38…
... Delay circuits, 16,17 ... Encoders, 18 ... Block distortion calculation circuits, 19 ... Code length memories, 20 ... Distortion memories, 26 ... Recording units, 27,28 ... Subtractors, 29 ... Reproduction unit, 32: Basic pixel memory, 33: Additional pixel memory, 36: Interpolation bottleneck, 37
…… Field memory, 39 …… Adder, 40 …… Decoder,
42 ... System controller.

Continuation of front page (58) Field surveyed (Int.Cl. ⁶ , DB name) H04N 7/24

Claims (1)

(57) [Claims] Input means for inputting image data, dividing means for dividing the image data input by the input means into blocks each comprising a plurality of pixels, and image data of each block representing basic information in the block, The first encoded data has a fixed number of bits and the second encoded data has a variable number of bits according to the block, representing information other than the basic information in the block. Image transmission, comprising: encoding means; and transmission means for arranging and transmitting the first encoded data encoded by the encoding means before the second encoded data. apparatus. 2. Image data is input, the input image data is divided into blocks each including a plurality of pixels, and the image data of each block represents basic information in the block, and has a fixed number of bits regardless of the block. 1 encoded data and second encoded data representing information other than the basic information in the block and having a variable number of bits according to the block, and the encoded first data The encoded data is stored in the second
An image transmission method characterized by arranging and transmitting the encoded data before the encoded data.