JPH08205130A - Image data transmission and reception device - Google Patents

Abstract

PURPOSE: To easily obtain a video of sufficient quality as a monitor image by transmitting a pseudo dynamic image by using a narrow-band transmission line and reproducing the pseudo dynamic image of specific image quality from the received image data like illustrations. CONSTITUTION: The video signal from a color camera is inputted to an image data transmission device 1 and separated 2 into R, G, and B and 8-bit quantized data are inputted to a register 3. Then a data comparison 6 between the register 3 and a 24-bit register 4 is made; when it is considered that both the registers have the same contents, a counted value is increased and the input number of the register 4 is counted. Then a counter value decision device 8 decides whether or not the counted value D of a counter 7 exceeds an upper-limit value; when data that can be regarded as the same data succeed, the counted value is made finite and a series of data compressing processings are performed in a constant time. Deleted RGB data and the counted value of the counter 7 are put together and sent to the narrow-band transmission line from the data transmission device 11, and the pseudo dynamic image of specific quality is reproduced from the received image data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image data transmitting / receiving apparatus for compressing and transmitting color image data and expanding and reproducing the received color image data.

[0002]

2. Description of the Related Art The frequency band of an analog public telephone network (telephone line) currently used is 0.3 to 3.4 KHz, and the transmission rate of a voice signal transmitted via a modem is usually It is 4.8 kbps (the transmission rate for data transmission is 9.6 kbps). Also, DDX, which is a digital communication network, has a maximum of 48
kbps is 64 kbps in the INS net 64 which is a digital telephone network (this transmission rate is 2 kHz of 3.4 KHz which is the maximum frequency of a voice signal according to the sampling theorem.
Double the 6.8 kHz, with a margin, the audio signal is sampled at 8 kHz and quantized with 8 bits, which is the signal transmission rate for one audio channel). ing.

However, the above-mentioned telephone network and communication network must be said to be extremely narrow band transmission lines for transmitting image data having a large amount of information. For example, a transmission rate of about 64 kbps to 1.5 Mbps is required in the fields of transmission of high-definition still images, videophones, video conferences, etc., and standard television broadcasting and HDTV (high definition)
Therefore, a transmission rate of about 1.5 Mbps to 156 Mbps is required. Furthermore, in the case of ultra-high definition TV, 15
A transmission rate of about 6 Mbps to 622 Mbps is required. Therefore, when transmitting various image data, a dedicated communication line (dedicated line) is currently used.

In addition, when performing communication at a transmission rate of 64 kbps to 1.5 Mbps, ISDN (Integrat
ed Services Digital Network, which is an integrated services digital communication network) is used.
This is a digital version of the telephone network,
Data communication, facsimile, still image transmission, videophone,
It is currently used in various communication fields such as video conferencing.
However, it cannot be used for a full-scale moving image (television image) or ultra-high-speed data communication because its transmission rate is too low. In order to realize a transmission rate of 1.5 Mbps or higher, B-ISDN (a broadband ISDN in which all the transmission lines are composed of optical fibers) is currently under development.

On the other hand, in an image data transmission system such as a fax machine, a videophone, a videoconference, etc., transmission is performed in consideration of the limitation of the bandwidth of the transmission line and the superposition of noise on the transmission line. The device is designed to reduce the amount of data to be processed as much as possible. For example, NTSC
(National Television System Committee) Standard 14
When transmitting 0 Mbps image data, the image data should be 1/1 as much as possible without degrading the image quality.
A data compression technique that compresses the data to 0 to 1/500 is required.

Therefore, in order to perform such data compression,
A data compression technique called high efficiency encoding technique is used. According to this technology, the redundancy existing in the image data to be transmitted is suppressed, and the amount of data is reduced so that the image quality is only slightly invisible to the eye. ing. In that case, when transmitting a binary image such as a FAX, a reversible encoding (which is a method of faithfully encoding the original image information as it is) is usually performed. When transmitting a moving image, lossy encoding (a method of encoding original image information without storing it as it is) is performed. This lossy encoding is intended to utilize the property that an image can be imaged in the human head to some extent.

In the case of performing such lossy coding, in order to efficiently use the correlation in the two-dimensional space possessed by the image information, in general, DCT (Discreat Cosine Transform) is used.
That is, a transform technique for data compression, which is a technique called a discrete cosine transform, has come to be used. In the DCT, in order to convert image information (that is, it can be treated as an aggregate of pixels on which the luminance signals of the three primary color signals of red, green, and blue are superimposed) from spatial coordinates to frequency coordinates, a predetermined pixel Number (usually 16 in the vertical direction x
For each block having 16 horizontal rows (256 in total), the target data string (usually, each pixel is 3 for red, green, and blue)
Each primary color signal is encoded as having 8-bit luminance information) is decomposed into a cosine waveform, and its decomposed component (corresponding to the frequency of the sequence Fourier transform) is extracted. Has become. And
It is possible to obtain a high coding rate by finely quantizing the low frequency component and roughly quantizing the high frequency component.

In this case, the high frequency component is visually rough, and deterioration is not noticeable even when quantized, and since the energy is originally small, the result of the quantization, that is, the value of the DCT coefficient can be made zero. It uses the property of high probability. By performing DCT conversion on the image data to be transmitted in this way, it becomes possible to greatly reduce the amount of data to be transmitted, and further, on the receiving side, by performing inverse DCT conversion on the received data. , The original image data can be reproduced.

The video signal (moving image) is generally 1
A still image of a frame is expressed as a continuous sequence, but this still image of one frame is a few frames per second.
A moving image is transmitted by transmitting 30 frames. Therefore, the above-mentioned DCT is also 1
This is performed for each still image of a frame (called intra-frame conversion coding), and in addition to this, the error signal predicted between two or more consecutive transmission frames is also D
CT is to be performed. This is called a hybrid DCT, but specifically, the image compression rate is further increased by predicting and extracting only the moving part of the screen. In this way,
It enables the transmission of moving images using a transmission path with band limitation.

On the other hand, regarding the image data compression technique,
International joint standardization work is in progress. The international standard encoding method for color still images is JPEG (Joint Pi
cture Image Coding Experts Group) is the international standard encoding method for color moving images, and MPEG (Mo
ving Picture Image Coding Experts Group) method,
Each is regulated. The JPEG method is a high-quality compression technology for still image data such as color photographs, jointly developed by ISO (International Organization for Standardization) and CCITT (International Telegraph and Telephone Advisory Committee).
It is a high-quality compression technology for color moving images that is being jointly developed by (International Organization for Standardization) and IEC (International Electrotechnical Commission). The DCT described above is adopted for any of these methods.

In the MPEG system, a technique for recording a color moving image in a storage system such as a CD-ROM is used as the MPEG-1 system, and as a MPEG-2 system, a high quality color image of a broadcasting station or a studio standard. Technologies for recording high-definition color images, such as HDTVs and HDTVs, are defined respectively. By the way, in order to compress image data using JPEG or MPEG and transmit a quasi-moving image, a relatively wideband transmission line such as the above-mentioned ISDN is required. In the field of videophones and videoconferencing, which are currently in practical use, such ISDN is utilized by taking advantage of the fact that the number of scene movements to be transmitted is very small on average.
Is used as a transmission path, and image data is transmitted at a high compression rate by JPEG or MPEG. There, the position is taken to allow a considerable amount of image distortion when a certain amount of motion occurs,
A transmission rate of 1.5 Mbps, which is 1/60 of a PCM (Puls Code Modulation) rate of about 90 Mbps, and an ISDN of 64 kbps, which is 1/1440.
The transmission rate that can be implemented in is used.

In ISDN, basically 64 kb
2 ps B channels (for signal transmission) and 16 kbp
One s s D channel (for control) is prepared. However, if high speed communication of 64 kbps or more is required, other 384 kbps H ₀ channel or 1.5
An H ₁ channel of 36 Mbps is prepared. Also,
In order to support ultra-high-speed data communication, TV moving images, HDTV, etc., the H _{2 to} H ₄ channels are also being studied. Therefore, in the conventional telephone line, at most 4.
What could only be transmitted at a transmission rate of 8 kbps or 9.6 kbps, but by using ISDN,
Higher speeds and higher quality communication capabilities have come to be realized, and various communication functions can be realized.

[0013]

However, in general, there are many image transmission systems that do not require such high quality images. For example, in a surveillance system for a warehouse or the like, in many cases, it is usually sufficient to simply transmit a still image as a surveillance image.

Therefore, it is never a good idea to apply the high quality image data transmission system using JPEG or MPEG as it is to such a surveillance system. Because in that case, I
It is necessary to lay a relatively high bandwidth transmission line such as SDN.
Further, it is necessary to use an expensive transceiver such as a codec. As a result, a problem in terms of cost when introducing the image transmission system will be greatly increased.

Therefore, a method is conceivable in which the compression rate of image data is extremely increased (for example, 1/20 or more) by JPEG or MPEG by using a telephone line or the like which is an inexpensive narrow band transmission line. . However, even if such a method is adopted, it is inevitable to use an expensive transmitter / receiver such as a codec, and since a narrow band transmission line is used, a blocking distortion appearing in a square shape on the screen. (That is, in the DCT, since encoding is performed for each pixel block, if there is a difference in image quality between the blocks, the image distortion of the blocks looks like a mosaic). , Mosquito noise generated in dots on the screen (that is, the quantization of the DCT coefficient is coarse, so the energy of the edge part included in the pixel block spreads into the block, and it looks like a mosquito pillar around the object. It is unavoidable that there is a problem with the phenomenon that various noises occur. As a result, the surveillance image becomes extremely difficult to see.

The present invention has been made to solve the above problem, and enables to transmit a quasi-moving image using an inexpensive narrow band transmission line, while a predetermined image data is received from the received image data. It is an object of the present invention to provide an image data transmitting / receiving apparatus capable of reproducing a quasi-moving image like an illustration with image quality.

[0017]

In order to achieve the above object, the present invention according to claim 1 is an image data transmitting / receiving apparatus for transmitting / receiving image data using a narrow band transmission line, It is composed of an image data transmitting device and an image data receiving device. Then, the image data transmission device
Sampling means for sampling the input image signal at equal intervals in time series, encoding means for encoding the sampled image signal into image data consisting of a plurality of bits, and a plurality of continuous image data, The data determining means for determining whether or not they can be substantially identified, and the first image data among the plurality of image data determined to be identifiable are defined by a predetermined ratio for each primary color signal of red, green, and blue. By combining the data amount reducing means for reducing the data amount, the first image data with the reduced data amount, and the data regarding the number of the remaining plural image data determined to be identifiable It is characterized by comprising: transmission data creating means for creating one piece of transmission data; and data sending means for sending the created transmission data in order to a narrow band transmission line. There.

Further, the image data receiving device includes a reception data sampling means for sampling the transmitted image data in order, and the sampled reception data.
Data separating means for separating the image data and the number data, and an image data sequence restoring means for restoring a sequence of continuous image data according to the separated image data and the number data,
It is characterized in that it is provided with a data interpolating means for interpolating between respective data in the restored image data sequence by the interpolation calculation, and an image data reproducing means for reproducing the image data according to the interpolated image data sequence.

Further, the invention according to claim 2 is the image data transmitting / receiving apparatus according to claim 1, wherein the encoding means converts the sampled image signal into red, green and blue primary color signals. It is characterized in that each of the separated primary color signals is quantized and encoded by a plurality of bits to form image data for each primary color signal. Also,
The invention according to claim 3 is the image data transmitting / receiving apparatus according to claim 1 or 2, wherein the data determining means further stores the first image data as a reference. Storage means, second storage means for sequentially storing second and subsequent image data to be compared with the first image data, and to change the reference image data,
The image data stored in the second storage means is converted into the first image data.
Data moving means for moving to the storage means, the image data stored in the first storage means, and the image data stored in the second storage means, the difference amount is obtained for each primary color signal, and each primary color signal is obtained. Data comparison means for making a comparison as to whether or not a reference predetermined value set for each signal is exceeded, and when all the obtained difference amounts for each primary color signal do not exceed the predetermined value, they can be identified as the same. Counting means for counting the number of times the data is repeated by incrementing the count value as image data, and determining whether or not the counting means counts the upper limit value of the count value If it is determined that the value has been counted, the count value is reset, the data moving means is operated, and the sampling means Is characterized by comprising a control means for controlling to resume pulling operation, the.

[0020]

According to the structure of the invention according to claim 1, the image data transmitting / receiving apparatus of the present invention transmits / receives image data using a narrow band transmission line such as a telephone line. More specifically, while transmitting a quasi-moving image using a narrow band transmission path, it is possible to reproduce the quasi-moving image in an illustrated manner with a predetermined image quality from the image data transmitted in this way. Has become.

Conventionally, as an image transmission system for transmitting a still image using a telephone line which is a narrow band transmission line,
Facsimile systems are well known. However, it takes 1 minute to transmit the contents of one A4 original, even for a G3 facsimile using an image compression technique called MR method, and the problem of transmission time has not been overcome. However, as a facsimile function, the problem of improving image quality is more important than the problem of transmission time.

On the other hand, if it is possible to transmit a quasi-moving image by using an inexpensive narrow band transmission line, it can be used in various fields, even if it is not a real-time image transmission like broadcasting. However, an example of a quasi-moving image transmission system using such a narrow band transmission line has not been found yet. This is because the quasi-moving image has an extremely large amount of data, and it is extremely difficult to transmit such a large amount of data through the narrow band transmission line.

However, by nature, an image is something that can be imagined to some extent in the human head, and normally, for example, as in a surveillance video system, almost only a still image is transmitted. In view of the fact that high image quality is not required so much in a field that can be used satisfactorily, the present inventors use an inexpensive narrow band transmission line and transmit a few still images in one minute. I wondered if it would be possible to transmit a quasi-video. Therefore, by configuring the image data transmitting / receiving apparatus as described above, the image data can be significantly compressed and transmitted, and when the compressed image data is received, the received data is subjected to interpolation processing. As a result, a smooth image can be reproduced.

More specifically, each of the above-mentioned constituent means operates as follows. First, in the image data transmitting apparatus, the image signal input from the television camera or the like is sampled at equal intervals in time series by the sampling means. Then, the encoding means encodes the sampled image signal into image data composed of a plurality of bits.

Next, the data judging means judges whether or not a plurality of consecutive image data can be substantially identified with each other. Therefore, for the first image data among the plurality of image data determined to be identifiable, the data amount reduction means reduces the data amount for each primary color signal of red, green, and blue at a predetermined rate. Done. Further, the first image data whose data amount has been reduced and the data regarding the number of the remaining plural image data determined to be identifiable are combined by the transmission data creating means. Then, the created transmission data is sequentially transmitted to the narrow band transmission line by the data transmission means.

On the other hand, in the image data receiving device, the received image data is sampled in order by the received data sampling means. Then, the data separation means separates the sampled reception data into image data and number data. Therefore, the image data sequence restoration means restores a sequence of continuous image data according to the separated image data and the number data. Further, the data interpolating means interpolates between the respective data in the restored image data sequence. Then, the image data reproducing means reproduces the image data according to the interpolated image data string.

According to the structure of the present invention as set forth in claim 2, in the encoding means, after the sampled image signal is separated into the respective primary color signals of red, green and blue, the separated primary color signals are separated. Is quantized and encoded with multiple bits. In this way, the sampled image signal is converted into image data for each primary color signal. Also,
According to the third aspect of the present invention, in the data determination means, first, the first storage means stores first reference image data. Further, the second storage means sequentially stores the second and subsequent image data to be compared with the first image data. Further, in order to change the reference image data, the second
The image data stored in the storage means is moved to the first storage means by the data moving means.

Subsequently, the data comparison means obtains the difference amount between the image data stored in the first storage means and the image data stored in the second storage means for each primary color signal. A comparison is made as to whether or not it exceeds a predetermined reference value set for each primary color signal. Therefore, when all the obtained difference amounts for each primary color signal do not exceed the predetermined value, the count value is incremented by the counting means as the image data that can be identified, and the number of times the data is repeated. Is counted.

Further, the control means determines whether or not the counting means has counted the upper limit value of the count value. When it is determined that the upper limit value has been counted, the control means resets the count value and controls so that the sampling operation of the data moving means and the sampling means is restarted.

[0030]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be specifically described below with reference to the drawings. FIG. 1 is a block diagram showing a configuration of an image data transmitting apparatus in an image data transmitting / receiving apparatus according to the present invention. This image data transmission device quantizes and encodes a video signal of a quasi-moving image such as a surveillance image input from a color television camera (not shown), then compresses the data,
It has the function of sending to a telephone line, which is a narrow band transmission line.

As for 1, 1 frame is, for example, 1 in the vertical direction.
A video signal composed of 20 pixels in the horizontal direction and 160 pixels in the horizontal direction (note that the pixel configuration of one frame complies with the aspect ratio 3: 4 of the NTSC system, and in the case of a normal television broadcast signal, the vertical direction is 480 × horizontal direction 640
However, in this example, the video signal is time-sequentially distributed at intervals of one pixel at equal intervals in the horizontal and vertical directions. It is designed to be picked up. Therefore, in this sampling apparatus 1, the number of pixels to be transmitted is first reduced to 1/4, and finally 4800 (12
0 × 160/4 = 4800) pixels will be sampled.

Regarding the sampling rate at this time, the resolution and the field of use of the color television camera are sufficient for the illustration transmission of a quasi-moving image such as a surveillance image, and the data compression to be executed. Is determined in consideration of the CPU processing time and the like.
For example, when it is assumed that 5 frames are transmitted per second (in the case of a normal TV broadcast signal, 30 frames per second are used).
Frames are transmitted, but in the case of surveillance video, it is sufficient to transmit almost still video, so by transmitting about 5 frames per second, it is possible to transmit as a semi-moving picture in an illustration.) , 24 kHz (4800 × 5 = 24
000).

For each pixel sampled here, a red signal (signal amplitude R) obtained by a CCD image pickup device of three primary colors provided in a color television camera.
And the green signal (signal amplitude G) and the blue signal (signal amplitude B) are the luminance signal (signal amplitude Y) in a state of being superimposed on each other. 2 is to separate 4800 pixel data sampled by the sampling device 1 into a red signal (R), a green signal (G) and a blue signal (B), and quantize each color signal with 8 bits. Data RGB
It is a separation device. In this case, the following three relational expressions 1 to 3 shown in [Equation 1] are applied between the luminance signal and each color signal. Therefore, the RGB values are obtained based on the luminance information about the individual pixels obtained by sampling.

[Equation 1] Y = 0.30R + 0.59G + 0.11B ... RY = 0.70R-0.59G-0.11B ... B-Y = 0.30R-0 .59G + 0.89B ... Next, the obtained R, G, and B color signals are quantized with 8 bits by a quantizer (not shown). As a result, each color signal is converted into RGB data each having a brightness expression of 256 gradations. Note that the number of quantization bits is 8 here, but in practice, it is possible to freely select from 6 to 10 bits according to the required image quality.

In this way, the brightness information of one pixel is converted into 24-bit image data consisting of a combination of RGB data. And at this stage,
The amount of data to be transmitted is 115.2k per frame
It is a bit (24 × 4800 = 115200). Here, as described above, when 5 frames are transmitted per second, 576 kbps (115.
A transmission rate of 2 × 5 = 576) is required.

However, this transmission rate is still 3.
It is not at a level that can be transmitted by a 4 kHz telephone line. Therefore, the present apparatus further reduces the data amount of the image data to be transmitted of 115.2 kbit per frame obtained by the data RGB separation device 2 by adopting the device configuration described below. I am supposed to do it.

Reference numeral 3 is a 24-bit first register that individually holds RGB data quantized to 8 bits for one pixel, and 4 is RGB data quantized to 8 bits for the next subsequent pixel. It is a 24-bit second register that holds each individually. Also, 5 is the second register 4
Is a data moving device for moving the RGB data held in the first register 3 to the first register 3.

Reference numeral 6 denotes RGB stored in the first register 3.
The data comparison device compares the data value and the RGB data value held in the second register 4 for each RGB data. Here, with respect to two consecutive pixels, it is detected whether or not the subsequent pixel is visually in a permissible range in which the change can be ignored with respect to the previous pixel. Note that, although this allowable range is experimentally obtained, for example, when the change amount of each of the RGB data is smaller than about 1/10 of 256 gradations, the permissible range is the same. It is considered to be in range.

Reference numeral 7 denotes a 4-bit counter, which counts the count value each time the data comparison device 6 detects that the amount of change in consecutive pixels does not exceed the allowable range for all RGB data. It is designed to increment. That is, the count value here indicates the number of consecutive pixels that can be regarded as substantially the same data.

Reference numeral 8 is a counter value judging device for judging whether or not the content of the counter 7 is smaller than a preset value. Here, an upper limit value is set for the count value of the counter 7, and the sampling of the sampling device 1 is continued until the count value reaches the upper limit value. That is, while the counting operation of the counter 7 is being performed, the image data of the pixels sampled continuously can be regarded as substantially the same data. After that, when the count value of the counter 7 reaches this upper limit value, the count value (upper limit value) at that time is given to the next data reduction device 9.

Reference numeral 9 denotes the data comparison device 6 in the case where the RGB data stored in the second register 4 cannot be identified with the RGB data stored in the first register 3 (that is, all of the RGB data of the next pixel). Exceeds the allowable range), or when the count value (upper limit value) is received from the counter value determination device 8, the RGB data held in the first register 3 is fetched and blue (B)> red. This is a data reduction device that reduces bits by the bit reduction amount of (R)> green (G).

As shown by the relational expression Y = 0.30R + 0.59G + 0.11B in the above-mentioned [Equation 1], this results in the blue (B) signal among the RGB signals forming the luminance signal Y. This is because it has the least impact. That is, the number of bits of the blue (B) signal can be reduced more than the others. In addition, each R
The actual bit reduction amount for the GB signal is obtained experimentally. In that case, R, which is the ratio of the RGB signals forming the luminance signal Y in the NTSC system, is:
Note that unless the relationship of G: B = 3; 6; 1 is not largely deviated, there is a possibility that hue changes that cannot be ignored can occur.

A data synthesizer 10 synthesizes the RGB data reduced by the data reducer 9 and the count value of the counter 7 as transmission data. The data reducer 9 is held in the first register 3. When the bits of RGB data are reduced, the data combining operation is performed. Therefore, for example, the counter 7
When the upper limit value of the count value of 14 is set to 14, and when the count value of 14 is received from the counter value determination device 8, the decimal number is the place where 336 (24 × 14 = 336) bits should be originally transmitted. The count value "14" at 2 is 2
Since it can be expressed by the decimal number "1110", that is, 4 bits,
In the end, only 24 bits + 4 bits = 28 bits need to be transmitted, and the data amount can be significantly reduced.

Therefore, if it is assumed that the above-described data compression is performed on all of the 4800 pixels constituting one frame described above (that is, 15 pixel data can be regarded as the same). When transmitting 5 frames per second, the transmission rate of 576 kbps is essentially required as described above, but the number of pixels is reduced to 320 (4800 ÷ 15 = 320) per frame. Therefore, the transmission rate is 44.8 kb
It is ps (28 × 320 × 5 = 44800). This transmission rate can be realized by using a voice band modem (a modem that uses a telephone line of 3.4 KHz and is a device used for converting a digital signal into a voice band and transmitting it). . Specifically, a transmission rate of 48 kbps will be realized by using a bandwidth of 48 kHz for voice channels. Therefore, by using this device, through a narrow band transmission line such as a telephone line,
It becomes possible to transmit the quasi-moving image data.

Reference numeral 11 is a data transmission device configured as such a modem. Here, the transmission data synthesized by the data synthesizing device 10 is transmitted to the communication opposite device via a telephone line or the like. In this case, the combined transmission data is code-converted to make the transmission data suitable for the transmission path, that is, an error correction code bit for coping with an error occurrence on the transmission path, and the receiving side. A horizontal sync bit and a vertical sync bit necessary for image reproduction at 1 are added.

A counter reset device 12 resets the contents of the counter 7 to zero when the transmission of one combined image data is completed. Then, at the same time as the counter 7 is reset, the next sampling is performed by the sampling device 1. FIG. 2 is a flowchart showing an operation process of the image data transmitting apparatus shown in FIG. It should be noted that here, a flowchart of the operation performed when one frame of image data is transmitted is shown. Therefore, when a plurality of frames are continuously transmitted, the contents of the flowchart shown here are repeated.

First, the sampling device 1 captures a video signal (for example, a video signal having a resolution of vertical pixels 120 × horizontal pixels 160) sent from a color television camera. In this case, the data RGB separation device 2 is used. hand,
The luminance signal of the first pixel in the first line of the video, which is the first data of the image frame, is separated into three primary color signals of red, green, and blue (hereinafter referred to as RGB separation), and each signal is 8 The signal is quantized, and each signal data is put into a predetermined position of the 24-bit first register 3 (S1).

Subsequently, the sampling device 1 starts sampling every other pixel in the horizontal and vertical directions of the video signals continuously input (S2). Further, the sampling data is sequentially separated into RGB by the data RGB separation device 2, and each signal is quantized by 8 bits (S
3). Then, the quantized RGB data are converted into 24
It is put in a predetermined position of the bit second register 4 (S4).

Next, the data comparison device 6 compares the data content stored in the first register 3 with the data content stored in the second register 4 (S5). Here, the first
The red, green, and blue color signal amounts stored in the register 3 are set to R1, G1, and B1, respectively, and the red, green, and blue color signal amounts stored in the second register 4 are set to R2, G2, and B2, respectively.
Further, assuming that the permissible range of each color signal amount is constants A, B, and C, | R1-R2 | <A, and | G1-G2 | <
B and | B1−B2 | <C are detected.

Then, as a result of this detection, when all of the above three comparison expressions are affirmed (Yes in S5)
Causes the 4-bit counter 7 to count one (S
6). In this process, when the data stored in the first register 3 and the data stored in the second register 4 can be identified with each other, the count value is incremented to sequentially enter the second register 4. The number of data (the number of repetitions) is counted.

Then, in the counter value judging device 8, the count value D of the counter 7 is the upper limit value (here, 15).
It is supposed that a determination is made as to whether or not it exceeds (S7). This processing is performed by making the count value finite when the data that can be identified are consecutive.
It is performed in order to complete the time required for a series of data compression processing within a fixed time. Then, as a result of the determination by the counter value determination device 8, when it is determined that the count value D is smaller than 15 (Yes in S7), the process returns to the sampling process of step S2, and the count value D is If the number exceeds 15 (No in step S7),
The process proceeds to step S8.

On the other hand, if all three comparison expressions are denied in the process of step S5 (No in S5), the data contents stored in the first register 3 and the second register 4 cannot be identified. Since this is the case, step S8
Move to the processing of. In step S8, in the data reduction device 6, the data stored in the first register 3 is taken out, the red signal (R) is changed from 8 bits to 4 bits, the green signal (G) is changed from 8 bits to 6 bits, and the blue signal is changed. (B) 8
Bits are reduced from 2 bits to 2 bits, respectively. By this processing, each 24-bit (8 + 8 + 8 = 24) RGB data is reduced to half 12 bits (4 + 6 + 2 = 12). Although the bit reduction amount in this case is obtained experimentally, as described above, N
It is preferable that the ratio is as close as possible to R: G: B = 3: 6: 1 in the TSC system, and by doing so, a more natural image can be reproduced at the time of reproduction.

Then, the data synthesizing device 10 synthesizes the RGB data reduced in step S8 and the count value of the counter 7 (S9). Note that step S
In the case of No in 7, in this example, 14 which is the upper limit value of the count value is combined with the bit-reduced data. Further, the combined data is converted by the data transmitter 11 into a code suitable for the transmission path and sent to a narrow band transmission path such as a telephone line (S10). Then, the data moving device 5 moves the data held in the second register 4 to the first register 3, and the counter resetting device 12 resets the contents of the counter 7 to zero (S11). It is prepared to process sampling data.

Finally, it is detected whether or not the end of the image frame is detected (S12), and when it is detected that the end of the image frame is detected (Yes in S12), the above series of processing is performed. End, otherwise (S12
If Yes), the process returns to the sampling process of step S2. The start and end of the image frame can be detected, for example, as follows. That is, the video signal input to the sampling device 1 has a slight no signal period after the horizontal synchronization signal provided in the previous stage when the video signal of one line composed of effective pixels starts. Therefore, by detecting the first signalless period, the beginning of the image frame can be confirmed, and sampling is started thereafter. Also,
The end of the image frame can be confirmed by detecting the value by the sampling device 1 because the total sampling number for one frame composed of effective pixels is a constant value. .

By the series of processing described above, for example, when 14 pieces of identifiable data are consecutive, 3 of the original data are used.
Since 36 bits (24 × 14 = 336) are reduced to 16 bits (24 ÷ 2 + 4 = 16), the reduction ratio is 16/336 = 1/21.
Therefore, it is possible to obtain image quality that is almost the same as the compression rate in the case of the conventional JPEG system.

Further, like the conventional JPEG system or MPEG system, a two-dimensional DCT (Discreat Cosine Transform)
When image data is transmitted using a narrow band transmission line using the above technology, frequency conversion is performed for each of a plurality of pixel blocks and the values are replaced with preset values. Although it was unavoidable to generate noise, in this device, in the processing of step S5 described above, data transmission is performed only when a certain variation amount is exceeded, so a low-pass filter is applied to the video signal. This is the same as in the above case, the occurrence of mosquito noise is eliminated, and since the blocking processing is not performed, the occurrence of blocking distortion is eliminated and the image quality can be improved.

FIG. 3 is a block diagram showing the configuration of the image data receiving device in the image data transmitting / receiving device according to the present invention. As shown in the figure, this image data receiving device is a reception data for sampling the image data continuously transmitted from the image data transmitting device shown in FIG. 1 through a narrow band transmission line such as a telephone line in units of 16 bits. A sampling device 21, a data separating device 22 for separating the sampled 16-bit image data into 12-bit RGB data and 4-bit number data (which is the number of RGB data equated), and the number data Based on the data restoration device 23 that restores a continuous image data sequence for each color signal of the red signal (R), the green signal (G), and the blue signal (B), and the restored image data sequence for each color signal. A three-point data interpolating device 24 that performs three-point data interpolation, and a display device 25 that reproduces and displays a quasi-moving image based on an image data string obtained as a result of the interpolation. Are al configuration.

Regarding the detection of the start and end of the frame on the receiving side, the start of the frame can be confirmed by detecting the no-signal period indicating the start of the frame, as described above. Further, the reception data sampling device 21 can confirm the end of a frame by detecting the total number of samplings (which is a constant value) for one frame to be transmitted.

FIG. 4 is a flow chart showing an operation process performed by the image data receiving apparatus shown in FIG. First,
The reception data sampling device 21 samples the image data continuously transmitted through a narrow band transmission line such as a telephone line in units of 16 bits (S21). Then, in the data separation device 22, 12 bits are a red signal,
Data separation is performed as image data for each color signal of the green signal and the blue signal, and the remaining 4 bits are the number of times the same data is repeated (or the number of the same data) (S22).

Further, the data restoration device 23 restores a continuous image data sequence based on the number of times the data is repeated (S23). Then, in the 3-point data interpolator 24,
A quadratic polynomial is created for each restored image data sequence of red, green, and blue signals, and interpolation is performed between three consecutive data (S24). The interpolation process in this case is performed as follows. For example, in the case of a red signal, the signal at the first point is Red (N) = 15, and the signal at the second point is Red.
(N + 1) = 9, Red (N + 2) = 1 for the third signal
And Then, a quadratic polynomial Red = a × N × N + b
For xN + c, N = 1, Red = 15, N = 2, Re
Substituting Red = 1 with d = 9 and N = 3, the coefficients a, b,
When c is obtained, a = -1, b = -3, and c = 19 are obtained. Here, a quadratic polynomial of Red = −N × N−3N + 19 is obtained. Therefore, by using this polynomial, the value of N = 1.5 or N = 2.5 can be obtained.

In this way, although the approximation processing is performed by sequentially performing the interpolation between the three consecutive data,
It is possible to obtain the value of the red signal for all pixels that compensate for the reduced pixels. Similarly, with respect to the green signal and the blue signal, the values for all the pixels can be obtained.
Therefore, the RGB data is simultaneously reproduced according to these values, and an image is displayed on the display device 25 (S25). In this case, the lack of image data due to data compression is covered by data interpolation, so that smooth gradation change can be expressed in pixel units.

In the above embodiment, the video signal input to the sampling device 1 in the image data transmitting device is set to 1
Sampling is performed every pixel, the number of bits of the counter 7 is 4 bits, and each color signal of red, green, and blue is reduced from 24 bits to 12 bits, and the signal ratio after reduction is 4: 6: 2. However, it goes without saying that the implementation conditions are not limited to all, and it is possible to set optimal conditions according to the system application, the required image quality, other communication conditions, and the like.

In the processing of received data in the image data receiving device, the number of sample points is 3 and 2
Interpolation is performed using the following polynomial, but of course, it is not limited to these conditions, and when a smoother gradation expression is required, a spline function may be used. Alternatively, interpolation may be performed with the number of sample points being four or more.

Further, the image data transmitting / receiving apparatus described in the above embodiment has an ordinary personal computer (for example, 80486 CPU manufactured by Intel Corporation, a clock frequency of 66).
It can be easily realized by mounting a video cap board on the one having a MHz microprocessor) and incorporating software having the contents shown in FIGS. Therefore, there is no need to use an expensive codec or the like as in the conventional case.

[0065]

As described above, according to the present invention, when an image data is transmitted using an inexpensive narrow band transmission line such as a telephone line, an image is formed using the conventional JPEG system or MPEG system. Without causing any problems such as blocking distortion and mosquito noise that occur when transmitting data, and without using expensive transceivers such as codecs at all, with a predetermined image quality, it is equivalent to the illustration. It becomes possible to transmit a moving image.

Therefore, when an image transmission system is used in a field such as a surveillance system of a warehouse, which is almost enough to transmit a still image in a normal case, the image data transmission / reception is not necessary. By using the device, it becomes possible to provide a video having an image quality that is sufficient as a surveillance video at an extremely low cost, and its practical effect becomes extremely large.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an image data transmitting apparatus in an image data transmitting / receiving apparatus according to the present invention.

FIG. 2 is a flowchart showing an operation process of the image data transmitting apparatus shown in FIG.

FIG. 3 is a block diagram showing a configuration of an image data receiving device in the image data transmitting / receiving device according to the present invention.

FIG. 4 is a flowchart showing an operation process of the image data receiving device shown in FIG.

[Explanation of symbols]

 1 Sampling Device 2 Data RGB Separation Device 3 First Register 4 Second Register 5 Data Moving Device 6 Data Comparison Device 7 Counter 8 Counter Value Judgment Device 9 Data Reduction Device 10 Data Synthesis Device 11 Data Transmission Device 12 Counter Reset Device 21 Received Data Sampling device 22 Data separation device 23 Data restoration device 24 Three-point data interpolation device 25 Display device

Claims (3)

[Claims] 1. An image data transmitting / receiving apparatus for transmitting / receiving image data using a narrow band transmission line, wherein the image data transmitting apparatus samples an input image signal in time series at equal intervals. Sampling means, encoding means for encoding the sampled image signal into image data composed of a plurality of bits, and data determining means for determining whether or not a plurality of consecutive image data can be substantially identified, The first of the plurality of image data determined to be visible
The second image data can be identified with the data amount reducing means for reducing the data amount at a predetermined rate for each of the red, green, and blue primary color signals and the first image data with the reduced data amount. Transmission data creating means for creating one transmission data by synthesizing the judged data of the plurality of remaining image data, and data sending for sending the created transmission data to the narrow band transmission line in order. While the image data receiving device is provided with a means, the image data receiving device, a reception data sampling means for sequentially sampling the transmitted image data, a data separation means for separating the sampled reception data into image data and number data, Image data sequence restoration means for restoring a sequence of continuous image data according to the separated image data and number data, and restoration by interpolation calculation An image data transmitting / receiving apparatus, comprising: a data interpolating unit that interpolates between respective data in the generated image data string; and an image data reproducing unit that reproduces image data according to the interpolated image data string. 2. The encoding means separates the sampled image signal into red, green, and blue primary color signals, quantizes each of the separated primary color signals, and encodes them with a plurality of bits, The image data transmitting / receiving apparatus according to claim 1, wherein the image data is provided for each primary color signal. 3. The data determination means further includes a first storage means for storing a first image data as a reference, and second and subsequent images to be compared with the first image data. Second storage means for sequentially storing the data; data moving means for moving the image data stored in the second storage means to the first storage means to change the reference image data; Image data stored by the first storage means;
Data comparing means for determining a difference amount from the image data stored by the second storage means for each primary color signal, and comparing whether or not the difference exceeds a predetermined reference value set for each primary color signal; Counting means for counting the number of times the data is repeated by incrementing the count value assuming that the obtained difference amounts for each primary color signal do not all exceed the predetermined value, assuming that the image data are identifiable image data. And, while the counting means determines whether or not the upper limit value of the count value is counted, when it is determined that the upper limit value is counted, the count value is reset and the data moving means is operated, Furthermore, the control means which performs the control which restarts the sampling operation of the said sampling means is provided, The claim 1 or 2 characterized by the above-mentioned. Image data transmission and reception apparatus Motomeko 2 wherein.