JPH0440123A - Picture data transmitting system - Google Patents

Abstract

PURPOSE:To establish synchronization even in the case of generating header error or bit error on the reception side by transmitting a packet while inverting one specified bit or plural bits in the packet at every fixed period, and generating a synchronous signal by detecting the inversion of the bits on the reception and reproduction side. CONSTITUTION:One specified bit in the packet is inverted as 1 and 0 at every fixed period. On the side of a receiving and reproducing device, synchronization is established by utilizing this bit inversion. Thus, even in the case of missing the packet or mixing the packet and generating error in a synchronization identification code, the restored picture is made normal after the next period.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention performs error-free synchronized reproduction and error detection and correction when transmitting encoded image signals in packets over an error-prone transmission path. The present invention relates to an image data transmission method that enables accurate transmission of image data.

[Summary of the invention]

The present invention relates to an image data transmission method that enables error-free synchronized reproduction and accurate error detection and correction when transmitting encoded image signals in packets over a transmission path with errors. Then, one specific bit or multiple bits in the packet are inverted and transmitted at regular intervals, and the receiving/reproducing side detects the inversion of the bit and generates a synchronization signal. When sequentially encoding and transmitting the difference values between pixel values or average values within a block of digitized image data, by encoding and transmitting the difference value with the previous data and the true value of the data for each synchronization position. , While ensuring synchronization on the receiving side, it is possible to minimize image degradation due to bit errors.

[Prior Art J] Various methods have been used to prevent errors when transmitting encoded image signals in packets through transmission paths where errors may occur or when storing them in packet format on storage media where errors may occur. However, in particular, one way to achieve synchronization when a bit error occurs in data is to insert a synchronization establishment code (marker code) at the position where synchronization is to be achieved in the encoded data, and the receiver side A method for detecting synchronization establishment codes and establishing synchronization is known.
No. 6 (1989)).

[Problem to be solved by the invention]

However, the conventional technology has the drawback that synchronization cannot be established when an error occurs in the synchronization establishment code itself or when a packet containing the synchronization establishment code is dropped due to a header error.

Therefore, in view of the above points, an object of the present invention is to be able to establish synchronization even if a header error or a bit error occurs on the receiving side, and to
An object of the present invention is to provide an image data transmission system that can prevent image deterioration due to bit errors.

[Means for Solving the Problems] The image data transmission method according to the present invention transmits a specific bit or multiple bits in the packet at regular intervals when transmitting an encoded image signal in a packet format. This is characterized in that the bit is inverted and transmitted, and the receiving/reproducing side detects the inversion of the bit and generates a synchronization signal.

Furthermore, the image data transmission method according to the present invention provides: - When sequentially encoding and transmitting difference values between pixel values or average values within a block of digitized image data to be transmitted, the difference value between the previous data and the previous data is The difference value of the data and the true value of the data are encoded and transmitted.

[Function] According to the present invention, one specific bit or a plurality of bits (hereinafter referred to as a synchronization identification code) in a packet of an encoded image signal are transmitted every fixed period (for example, every horizontal line of an image).
Invert it to .

Here, bit errors in packet data are: ■ packet loss/packet mix-up due to an error in the packet header;
0 errors in the image data itself, and when the above-mentioned packet loss or packet mix-up occurs, burst-like data loss or error data is added to the data, causing deviations and corruption in subsequent images. Furthermore, regarding errors in the image data itself, so-called out-of-synchronization occurs in encoded data using the variable length encoding method, and the errors spread.

Therefore, in the present invention, one or more specific bits in a packet are inverted (inverting 1 and 0) at regular intervals (for example, every horizontal line of an image), and on the receiver side, By detecting the bit reversal and resynchronizing, even if a data error occurs and the image in the relevant period (for example, the -horizontal line) is corrupted,
It is possible to prevent the error from spreading to subsequent stages.

Furthermore, by using such synchronization processing, it is possible for the receiver to detect that there is an error in the relevant period, and even more powerful error countermeasures such as interpolation with adjacent image data are possible. .

In addition, when sequentially encoding and transmitting the difference values between pixel values or average values within blocks of digitized image data to be transmitted, the difference value with the previous data and the true value of the data are encoded for each synchronization position. In the invention characterized in that the receiving side that has received the difference value and the true value compares the data value restored based on the difference value with the true value, thereby transmitting the data value. It is possible to detect the presence or absence of an error.

[Example] Examples of the present invention will be described in detail below.

Mitate Yu 1 to 4 show a first embodiment of the invention.

Here, FIG. 1 is a diagram showing a transmission packet format of encoded image data.

FIG. 2 shows a block configuration for sending out the packet shown in FIG. 1. In this embodiment, one specific bit in the packet is inverted to 1°0 at regular intervals (
(See Figure 1).

FIG. 3 shows a bitstream with only synchronization identification bits. The receiver/player side utilizes this bit inversion to establish synchronization.

Figure 4 shows when packet loss or packet mix-up occurs,
An example of a bitstream of a synchronization identification code when an error occurs in the synchronization identification code is also shown. As shown in FIG. 4(B), even if a packet is lost, an error will occur during the relevant period, but a normal restored image will be obtained from the next period onward. Furthermore, as shown in FIG. 4(C), even if packets are mixed, an error will occur in the same period, but
After the next period, the restored image becomes normal. Furthermore, Figure 4 (
As shown in D), even if an error occurs in the synchronization identification code, it is generally extremely rare for bit errors to occur continuously, so errors that occur sporadically as shown in this figure should be ignored. As a result, a normal restored image can be obtained including the period in which the error occurred.

husband's blood FIG. 5 is a flowchart showing the second embodiment.

The embodiment described in detail below uses a synchronization code such as that described above, detects on the receiver/player side that an error has occurred within a certain period, and performs error concealment by interpolation.

The operation of this embodiment is as follows.

■ In the case of an encoding method using equal-length codes (step S2
), the amount of data within a certain period is constant and the number of packets is constant.

Therefore, on the receiver side, by counting the number of synchronization identification codes (number of consecutive 1s or 0s) within a certain period (step S3), when there is an excess or deficiency (when a packet is dropped or a packet is mixed), (Step S4), it is detected that an error has occurred in the period, and stronger error countermeasures are taken by interpolating with data one frame before the period or adjacent data of the same frame (Step S5). I can do it.

■ In case of encoding method using variable length code (step S
Regarding 2), when transmission is performed using a variable length code, the amount of encoded data changes within a certain period, but the amount of decoded data (for example, the number of pixels) is constant, so the decoding The number of converted data is counted (step s6), and if there is an excess or deficiency (step S7), error concealment (step S8) becomes possible in the same manner as in the above (2).

Next, a third embodiment in which a DPCM code and a variable length code are combined will be described with reference to FIGS. 6 and 7.

Generally, as an encoding method for image data, variable length encoding is often performed after performing DPCM processing on pixel values and intra-block average values. , pixel data at the beginning of each line)
Regarding this, the PCM value is sent instead of the difference value, causing an error to spread. The situation is shown in FIG. 6(A). As shown in this figure, the difference value is not transmitted for the first data of each period.

In contrast to the transmission method shown in FIG. 6(A), the third method of the present invention
In this embodiment, as shown in FIG. 6(B), for the first data of each period, in addition to the PCM value, the difference value from the last data of the previous period is also transmitted. As a result, on the receiver side, if the value reproduced from the difference value and the PCM value do not match, it is determined that there was an error in the previous period, and error concealment is performed for the previous period as in the second embodiment. Do this.

FIG. 7 is a flowchart showing the operation of the third embodiment shown in FIG. 6(B).

〔Effect of the invention〕

As explained above, according to the present invention, synchronization can be established even if a header error or a bit error occurs.

Furthermore, according to the present invention, it is possible to extremely minimize (suppress) image deterioration due to bit errors with an extremely small increase in data amount.

In addition to the effects of the invention described above, effects unique to each embodiment can be obtained.

In the first and second embodiments described above,
The amount of increase in data for error countermeasures is one bit of the synchronization identification code per packet, which is extremely small. Also in the third embodiment, it is sufficient to transmit only one extra piece of difference value data.

Furthermore, the third embodiment can be combined with other synchronization establishment codes, and in that case, transmission and storage methods other than the packet method are also possible.

Ru.

2...Packet generation unit, 4...Synchronization identification code insertion section.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the packet format in the first embodiment of the present invention, FIG. 2 is a block configuration diagram for outputting a packet signal according to the first embodiment, and FIG. 3 is a diagram showing the packet format in the first embodiment. A diagram showing a pit stream of the synchronization identification code used; FIG. 4 is an explanatory diagram when various transmission errors occur in the first embodiment;

Claims (1)

[Claims] 1) When transmitting an encoded image signal in a packet format, a specific bit or multiple bits in the packet are inverted and transmitted at regular intervals, and the receiving/reproducing side An image data transmission method is characterized in that it detects the inversion of the relevant bit and generates a synchronization signal. 2) When sequentially encoding and transmitting the difference values between pixel values or average values within blocks of digitized image data to be transmitted, the difference value with the previous data and the true value of the data are encoded for each synchronization position. An image data transmission method that is characterized by converting the data into images and sending them out. 3) In claim 2, the receiving side that has received the difference value and the true value detects the presence or absence of a transmission error by comparing the data value restored based on the difference value with the true value. An image data transmission method characterized by: