Transmitting apparatus

BACKGROUND OF THE INVENTION

Field of the Invention

0001The present invention relates to a transmitting apparatus and, more particularly, to a transmitting apparatus in which information of another channel is inserted into information of one channel in a transmitting system for compressing information by a prediction coding and a non-prediction coding and transmitting the compressed information.

Related Background Art

0002A digital video tape recorder (VTR) is known as an image recording apparatus for digitally compressing a moving image and recording onto a recording medium such as a magnetic tape or the like. There are a fixed length coding compression method and a variable length coding compression method as compression methods. As compared with the fixed length coding compression method, in the variable length coding compression method, although a high compression ratio of about 1/10 to 1/20 can be accomplished without largely deteriorating a picture quality, there is a drawback such that a compression data amount per picture plane (frame or field) is not uniform and it is difficult to perform the edition (joint recording or the like), retrieval, and special reproduction (random access, fast forward reproduction, reverse reproduction, etc.).

0003An MPEG (Moving Picture Coding Expert Group) uses an intra-frame (field) coding and a prediction coding which uses a future and/or past frame (or field) as a reference picture plane and periodically arranges the intra-frame coded picture plane (hereinafter, called an I frame) and, on the other hand, sets the picture plane in the I frame to a prediction coded picture plane. According to such a recording method, the compression data of an inter-frame prediction coding picture plane in the reverse direction or both directions is recorded after the compression data of the prediction reference picture plane on the recording medium.

0004According to such a recording method, for instance, a limitation occurs at the joint position of the image, for instance, in the insertion recording (joint recording). For example, even if the operator tries to join another picture plane just after the inter-frame prediction coding picture plane in the reverse direction or both directions, the image data of the picture plane which is necessary to decode such an inter-frame prediction coding image in the reverse direction or both directions needs to be preserved on the recording medium. On the other hand, a prediction relation between the old picture plane to be inserted and the new picture plane to be inserted becomes discontinuous in the joint portion, so that a reproduction picture plane becomes disorder.

0005Such a problem is not limited to only the VTR as mentioned above but a similar problem also occurs in case of inserting data of another channel to a predetermined insertion point of data of one channel.

0006According to one aspect of the invention, there is provided an image processing apparatus for inserting second image data into first moving image data including intra-picture coded image data and bidirectionally predictable inter-picture coded image data, comprising: <ul id="ul0001" list-style="none" compact="compact"><li>a) inserting means for inserting the second moving image data at a selected position into the first moving image data to generate combined moving image data; and</li><li>b) control means for enabling the selected position at which the second moving image data are to be inserted into the first moving image data to be adjusted so that the position at which the second moving image data is actually inserted is such that when the combined moving image data is decoded, the first moving image data can be decoded up to at least the selected position,</li></ul>    wherein said control means is arranged to insert into the first moving image data a flag for causing image data of frames of the first moving image data occurring between the selected position and the actual insertion position of the second moving image data to be skipped or omitted during reproduction of the combined moving image data.

0007In another aspect, the present invention provides an image processing method of inserting second moving image data into first moving image data including intra-picture coded image data and bi-directionally predictable inter-picture coded image data, comprising: <ul id="ul0002" list-style="none" compact="compact"><li>a) an inserting step of inserting the second moving image data at a selected position into the first moving image data to generate combined moving image data; and</li><li>d) a control step of controlling the selected position at which the second moving image data are to be inserted into the first moving image data to be adjusted so that the second moving image data is inserted at an actual insertion position so that when the combined moving image data is decoded, the first moving image data can be decoded up to at least the selected position,</li></ul>    wherein said control step is arranged to insert into the first moving image data a flag for causing image data of frames of the first moving image data occurring between the selected position and the actual insertion position of the second moving image data to be skipped or omitted when the combined moving image data is read.

0008In an embodiment, a transmitting apparatus according to the present invention is a recording apparatus such as a VTR or the like and is a recording apparatus for recording new information from the designated position of the existing record in a recording system for compressing a series of information signals by a non-prediction coding and a prediction coding using non-prediction coding information as reference information and for recording the prediction coding information and the non-prediction coding information onto a recording medium in accordance with a predetermined order, wherein the recorded information, namely, the past non-prediction coding information using the non-prediction coding information just after the insertion point as reference information is erased, and a first predetermined number of information among information to be inserted are prediction coded by using the non-prediction coding information just before the insertion point of the recorded information as reference information.

0009By the above means, regularities of the prediction coding and the non-prediction coding can be maintained even by the insertion recording, and it is sufficient to execute the same reproducing process at time points before and after the insertion points. The invention can also easily cope with the edition, retrieval, special reproduction, or the like.

0010The above and other concerns and features of the present invention will become apparent from the following detailed description and the appended claims with reference to the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

0011<ul id="ul0003" list-style="none" compact="compact"><li>Figs. 1A and 1B, when combined together as shown Fig. 1, show a schematic constructional block diagram of an embodiment of the present invention;</li><li>Fig. 2 is a schematic constructional block diagram of a movement compensation frame prediction circuit 38;</li><li>Fig. 3 is a schematic constructional block diagram of a movement compensation frame prediction circuit 73;</li><li>Fig. 4 is an explanatory diagram of a prediction direction;</li><li>Figs 5A to 5E are timing charts for an example not within the scope of what is claimed;</li><li>Fig. 6 is an example of a data arrangement on recording track patterns corresponding to Figs. 5A to 5E;</li><li>Figs. 7A to 7F are timing charts for an insertion recording in the embodiment;</li><li>Fig. 8 is an example of a data arrangement on recording track patterns corresponding to Figs. 7A to 7F;</li><li>Figs. 9A to 9E are timing charts for a second example not within the scope of what is claimed;</li><li>Fig. 10 is an example of a data arrangement on recording track patterns corresponding to Figs. 9A to 9E.</li></ul>

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

0012An embodiment of the present invention will be described hereinbelow with reference to the drawings.

0013Figs. 1A and 1B show a transmitting apparatus of an embodiment of the present invention. The embodiment relates to the case where the invention was applied to a digital VTR.

0014In Figs. 1A and 1B, reference numeral 10 denotes an analog input terminal of a moving image signal to be recorded; 12 an A/D converter to convert the analog image signal from the input terminal 10 into the digital signal; 14 a block forming circuit to form blocks each comprising [(i) pixels in the horizontal direction x (j) pixels in the vertical direction] from the raster scan image data which is output from the A/D converter 12; 16 a digital input terminal of a moving image signal to be recorded; and 18 a buffer of the image data from the input terminal 16. It is now assumed that the image data which is supplied to the input terminal 16 has already been converted into the blocks. Reference numeral 20 denotes a switch to select an output of the block forming circuit 14 or an output of the buffer 18.

0015Reference numeral 22 denotes a subtractor to subtract a prediction value of an output of the switch 20; and 24 indicates a switch to select either one of the output (a contact) of the switch 20 and an output (b contact) of the subtractor 22. The switch 24 is connected to the (a) contact in case of the intra-frame coding and to the (b) contact in case of the inter-frame prediction coding.

0016Reference numeral 26 denotes a DCT (Discrete Cosine Transform) circuit to convert the data selected by the switch 24 into a frequency region by a discrete cosine transformation; and 28 a quantizing circuit to quantize an output (frequency coefficient) of the DCT circuit 26 by quantization steps of each frequency coefficient.

0017Reference numeral 30 denotes an inverse quantizing circuit for inverse quantizing an output of the quantizing circuit 28; 32 an inverse DCT circuit to convert an output of the inverse quantizing circuit 30 into a time base region; and 34 an adder for adding 0 to an output of the inverse DCT circuit 32 in case of the intra-frame coding and for adding the prediction value to the output of the inverse DCT circuit 32 in case of the inter-frame prediction coding. An output of the adder 34 becomes a local decoding value of the intra-frame coding or the inter-frame prediction coding.

0018Reference numeral 36 denotes a movement vector detecting circuit to detect a movement of an image between two picture planes (ordinarily, the present picture plane and the preceding picture plane); and 38 a movement compensation frame prediction circuit to movement compensate an output of the adder 34. An output of the movement compensation frame prediction circuit 38 becomes a prediction value. Reference numeral 40 denotes a switch which is opened in case of the intra-frame coding and is closed in case of the inter-frame coding. The prediction value which is output from the movement compensation frame prediction circuit 38 is supplied to the subtractor 22 and is also supplied to the adder 34 through the switch 40.

0019Reference numeral 42 denotes a variable length coding circuit for variable length coding the output of the quantizing circuit 28; and 44 a buffer memory for buffering an output of the variable length coding circuit 42. Quantizing characteristics of the quantizing circuit 28 are controlled in accordance with a memory data amount of the buffer 44. Reference numeral 46 denotes an addition circuit for adding an error correction code, a synchronizing code Sync, an ID signal, and a discrimination signal to discriminate the intra-frame coding and the inter-frame coding. Reference numeral 48 denotes a modulation circuit for low frequency suppression modulating an output of the addition circuit 46; and 50 a recording amplifier to amplify an output of the modulation circuit 48 to a predetermined level.

0020Reference numeral 52 denotes a magnetic head for recording and reproduction; and 54 a magnetic tape as a recording medium.

0021Reference numeral 56 denotes a reproducing amplifier to amplify a reproduction output of the recording and reproducing head 52; 58 a demodulation circuit to demodulate an output of the reproducing amplifier 56; 60 a data reproduction circuit for reproducing the data in accordance with the sync code Sync and the ID signal and performs an error correction by the error correction code; 62 a buffer memory for buffering a reproduction data which is output from the data reproduction circuit 60; 64 a variable length decoding circuit for variable length decoding output data of the buffer 62; 66 an inverse quantizing circuit for inverse quantizing an output of the variable length decoding circuit 64; and 68 an inverse DCT circuit for inverse DCT converting an output of the inverse quantizing circuit 66.

0022Reference numeral 70 denotes an adder for adding a prediction value to an output of the inverse DCT circuit 68; and 72 a switch to select either one of an output (<u>a</u> contact) of the inverse DCT circuit 68 and an output (b contact) of the adder 70. In accordance with the coding discrimination signal detected by the data reproduction circuit 60, the switch 72 is connected to the (a) contact in case of the intra-frame coding and to the (b) contact in case of the inter-frame coding.

0023Reference numeral 73 denotes a movement compensation frame prediction circuit to calculate a movement compensation frame prediction value from the reproduction image data selected by the switch 68. Although the same circuit as the movement compensation frame prediction circuit 38 of the recording system is commonly used as a movement compensation frame prediction circuit 73, the circuit 73 is indicated by a different reference numeral for convenience of explanation of the recording and reproducing operations.

0024Reference numeral 74 denotes an inverse block forming circuit to return the reproduction image data selected by the switch 72 to the raster scan data; 76 a D/A converter for converting output data of the inverse block forming circuit 74 into the analog signal; 78 an analog reproduction image output terminal to generate an output of the D/A converter 76 to the outside; 80 a buffer for buffering the reproduction image data selected by the switch 72; and 82 a digital reproduction image output terminal to generate an output of the buffer 80 to the outside.

0025Reference numeral 84 denotes a switch for transferring the image data which is necessary for the inter-frame coding of the image to be inserted in the recorded image in case of the insertion recording to the movement compensation frame prediction circuit 38. Reference numeral 86 denotes a control circuit to control a whole apparatus, particularly, control the switching operations of the switches 20, 24, 40, and 84.

0026Fig. 2 shows a detailed circuit diagram of the movement compensation frame prediction circuit 38. Fig. 3 shows a detailed circuit diagram of the movement compensation frame prediction circuit 73. Each of the circuits 38 and 73 comprises: two frame memories 100 and 102 which are serially connected; a front direction prediction circuit 104; a back direction prediction circuit 106; an adder 108 to add an output of the front direction prediction circuit 104 and an output of the back direction prediction circuit 106; and a switch 110 for selecting either one of the outputs of the front direction prediction circuit 104, back direction prediction circuit 106, and adder 108.

0027As shown in Fig. 2, in the recording system, the movement vector detected by the movement vector detecting circuit 36 is supplied to the front direction prediction circuit 104 and back direction prediction circuit 106. The front and back direction prediction circuits 104 and 106 output prediction values which were movement compensated with respect to the front and back directions, respectively.

0028The switching operation of the switch 110 is controlled by the control circuit 86 in accordance with a compression effect, a picture quality, a special reproducing function, or the like.

0029Fig. 4 shows an example of the recording order of an intra-frame coding picture plane (hereinafter, called an I frame), a one-side prediction coding picture plane (hereinafter, called a P frame) using one future or past I frame as a reference picture plane, and a both-side prediction coding picture plane (hereinafter, called a B frame) in which a both-side prediction was performed by using the I and P frames as reference picture planes. Appearance frequencies and orders of the I, P, and B frames are adaptively adjusted in accordance with a compression ratio, a picture quality, or the like.

0030The ordinary recording and reproducing operations will be first described. In the ordinary recording and reproducing operations, the control circuit 86 turns off the switch 84.

0031In case of the analog input, the A/D converter 12 converts the analog image signal from the input terminal 10 into the digital signal. The block forming circuit 14 forms blocks from the output image data of the A/D converter 12. The switch 20 is connected to the (a) contact. In case of the digital input, the input image data of the input terminal 16 is supplied to the (b) contact of the switch 20 through the buffer 18. The switch 20 is connected to the (b) contact.

0032In case of the picture plane which is intra-frame coded, the control circuit 86 connects the switch 24 to the (a) contact and turns off the switch 40. Thus, the output (output of the block forming circuit 14 or output of the buffer 18) of the switch 20 is DCT converted by the DCT circuit 26 and is quantized by the quantizing circuit 28.

0033The output of the quantizing circuit 28 is decoded by the inverse quantizing circuit 30 and the inverse DCT circuit 32. The decoded value passes through the adder 34 and is supplied to the movement vector detecting circuit 36 and the movement compensation frame prediction circuit 38. The movement vector detecting circuit 36 detects a movement vector of the image from the output of the adder 34 and the present image data from the switch 20 and supplies to the movement compensation frame prediction circuit 38. In the movement compensation frame prediction circuit 38, the output of the adder 34 is sequentially stored into the frame memories 100 and 102 and generates a prediction value which was movement compensated in accordance with the movement vector detected by the movement vector detecting circuit 36.

0034In case of the picture plane which is inter-frame coded, the control circuit 86 connects the switch 24 to the (b) contact and turns on the switch 40. The subtractor 22 subtracts the prediction value from the movement compensation frame prediction circuit 38 from the output of the switch 20 and generates a prediction error. The prediction error is supplied to the DCT circuit 26 through the switch 24 and is DCT converted by the DCT circuit 26 and is quantized by the quantizing circuit 28.

0035The output of the quantizing circuit 28 is decoded by the inverse quantizing circuit 30 and the inverse DCT circuit 32. The adder 34 adds the prediction value to the output of the inverse DCT circuit 32. The output of the adder 34 is a local decoded value of the inter-frame coded data. The output of the adder 34 is supplied to the movement compensation frame prediction circuit 38 but is not stored into the frame memories 100 and 102.

0036The data which was intra-frame coded or was inter-frame coded as mentioned above is variable length coded by the variable length coding circuit 42. The output of the circuit 42 is supplied to the addition circuit 46 through the buffer 44 and is added with the error correction code, sync code Sync for synchronization block formation, ID signal for synchronization block discrimination, and coding discrimination signal and is generated as a sync block of a predetermined size. The coding discrimination signal is supplied from the control circuit 86.

0037The modulation circuit 48 low frequency suppression modulates the output of the addition circuit 46. The output of the modulation circuit 48 is supplied to the magnetic head 52 through the recording amplifier 50 and is recorded onto the magnetic tape 54.

0038In the ordinary reproducing mode, the recording signal of the magnetic tape 54 is reproduced by the magnetic head 52 and its head output is supplied to the demodulation circuit 58 through the reproducing amplifier 56 and is demodulated. The data reproduction circuit 60 detects the ID, compression data, error correction code, and coding discrimination signal from the output of the demodulation circuit 58 by the sync code Sync and corrects a recording and reproduction error by the error correction code. The switching operation of the switch 72 is controlled by the detected coding discrimination signal.

0039The output (reproduction data) of the data reproducing circuit 60 is supplied to the variable length decoding circuit 64 through the buffer 62. The variable length decoding circuit 64 decodes the variable length code portion in the compression data and its output is inverse quantized by the inverse quantizing circuit 66 and is inverse DCT converted by the inverse DCT circuit 68.

0040The adder 70 adds the prediction value to the output of the inverse DCT circuit 68. The output of the inverse DCT circuit 68 is supplied to the (a) contact of the switch 72. The output of the adder 70 is supplied to the (b) contact of the switch 72. Under control of the data reproduction circuit 60, the switch 72 is connected to the (a) contact for the intra-frame coding frame and to the (b) contact for the inter-frame coding frame. Therefore, the output of the switch 72 becomes reproduction image data in which the intra-frame coding data and the inter-frame coding data were respectively properly decoding processed.

0041The movement compensation frame prediction circuit 73 generates a prediction value of the movement compensation frame prediction from the output of the switch 72 in a manner similar to the movement compensation frame prediction circuit 38.

0042The inverse block forming circuit 74 returns the output of the switch 72 to the raster scan output. The output of the inverse block forming circuit 74 is converted into the analog signal by the D/A converter 76 and is output from the analog output terminal 78 to the outside. The output of the switch 72 is also rate adjusted by the buffer memory 76 and is output from the digital output terminal 82 to the outside.

0043The operation of the insertion recording (joint recording) for recording so as to join a new video image from the halfway of the existing recorded video image will now be described with reference to Figs. 5A to 5E. It is now assumed that the image to be inserted is input from the input terminal 10 or 16. For simplicity of explanation, however, it is assumed that the inter-frame prediction coding picture plane is only the B frame using the past and future intra-frame coding picture planes (I frames) as reference picture planes. The coding data of the intra-frame coding picture plane is shown by adding (i) after the frame number. The coding data of the inter-frame coding picture plane is shown by adding b(k1, k2) after the frame number. k1 and k2 denote frame numbers of the reference picture planes.

0044Fig. 5A shows a prediction relation between the inputting order (or reproducing order) of a video image signal to be insertion edited and the inter-frame coding. Fig. 5B shows a recording order (coding order). Fig. 5A corresponds to the output of the switch 20. Fig. 5C shows an inputting order of a video image signal to be inserted. Fig. 5D shows a prediction relation and a coding method according to the embodiment of a video image signal to be inserted. Fig. 5E shows a recording order (coding order) after completion of the insertion.

0045It is now assumed that images of the frame #50 and subsequent frames shown in Fig. 5C are insertion recorded after the frame #3. In this case, since the frame #3 has been both-side inter-frame coded by using the frames #1 and #4 as reference frames, coding data I4 of the frame #4 cannot be erased.

0046In this example outside the scope of what is claimed, therefore, compression data 4i of the frame #4 which is necessary for decoding of the frame #3 is left and the recording of the images to be inserted (frame #50 and subsequent frames) is started from the position at which the intra-frame coding picture plane should be arranged after 3b(0,4).

0047In case of the images to be inserted (frame #50 and subsequent frames) as they are, the frames #50 and #54 become the I frames and the frames #51, #52, and #53 become the B frames using the frames #50 and #54 as reference frames. However, in order to maintain the period of the I frame, in the example, in place of the frame #50, the frame #2 is used as reference frames of the frames #51, #52, and #53.

0048In the process for retrieving the recording start point of the insertion recording, the switch 84 has been turned on and the reproduction data of the I frame (in the example, frame #4) just after the insertion point of the existing record is transferred to the movement compensation inter-frame prediction circuit 38 of the recording system. Thus, the frame #4 can be used as a reference frame at the time of the inter-frame prediction coding of the frames #51, #52, and #53 of the images to be inserted.

0049In case of commonly using the movement compensation inter-frame prediction circuit 38 of the recording system and the movement compensation inter-frame prediction circuit 73 of the reproducing system, when the insertion recording is executed, it is sufficient to preserve the memory contents in the frame memories 100 and 102.

0050In the example, therefore, the frames #51, #52, and #53 are inter-frame prediction coded by using the frames #4 and #54 as reference frames. After completion of the insertion recording, the recording order of the coding data on the magnetic tape 54 is as shown in Fig. 5E.

0051By recording as mentioned above, not only the arrangement rules of the intra-frame coding and the inter-frame coding are maintained at time points before and after the insertion but also the relation of the reference frames of the inter-frame prediction is maintained, so that the reproducing operation can be simplified and a burden on the reproducing process is reduced.

0052By recording as shown in Fig. 5E, as a reproduction output order, the frame #51 and subsequent frames are reproduced and output after the frames #1, #2, #3, and #4. Namely, this equivalents to that the frames #51 and subsequent frames were insertion recorded after the frame #4. However, a deviation of one frame doesn't cause a large problem.

0053Fig. 6 shows an example of recording track patterns according to the example.

0054An embodiment of the invention, in which the I trame after the insertion point of the recorded image is preserved and the first frame of an image to be inserted is not deleted, will now be described. Fig. 7A shows a prediction relation between the inputting order (or reproducing order) of the video image signal to be insertion edited and the inter-frame coding. Fig. 7B shows a recording order (coding order). Fig. 7C shows an inputting order of the video image signal to be inserted. Fig. 7D shows a prediction relation and a coding method according to the embodiment of the video image signal to be inserted. Fig. 7E shows a recording order (coding order) after completion of the insertion. Fig. 7F shows a reproduction picture plane order of the recorded image after completion of the insertion.

0055It is now assumed that the frame #50 and subsequent frames are inserted just before the frame #3. In this case, although the frames #3 and #4 are unnecessary, since the frame #2 has already been both-side inter-frame coded by using the frames #1 and #4 as reference frames, the coding data 4i of the frame #4 cannot be erased.

0056Therefore, in the embodiment, the compression data 4i of the frame #4 which is necessary for decoding of the frame #2 is left and data 3b(0, 4) is also left in order to maintain the regularity. An insertion flag indicative of the actual insertion point is recorded after the data 4i and the images (frame #50 and subsequent frames) to be inserted are recorded after the data 3b(0, 4). For the images (frame #50 and subsequent frames) to be inserted, the frame #4 is set to the first picture plane and the frames are subsequently sequentially coded in accordance with a predetermined order. That is, the frames #50, #51, and #52 are inter-frame coded by using the frames #4 and #53 as reference frames.

0057In a manner similar to the case of Figs. 5A to 5E, for the process to retrieve the recording start point of the insertion recording, the switch 84 has already been turned on and the reproduction data of the I frame (frame #4 in the embodiment) just after the insertion point of the existing record is transferred to the movement compensation inter-frame prediction circuit 38 of the recording system. Due to this, the frame #4 can be set to the reference frame at the time of the inter-frame prediction coding of the frames #50, #51, and #52 of the image to be inserted.

0058In the case where the movement compensation inter-frame prediction circuit 38 of the recording system and the movement compensation inter-frame prediction circuit 73 of the reproducing system are commonly used, when the insertion recording is executed, it is sufficient to preserve the memory contents in the frame memories 100 and 102.

0059After completion of the insertion recording, the recording order of the coding data on the magnetic tape 54 is as shown in Fig. 7E. When the coding data is reproduced, the frame #3 is read-skipped by the insertion flag and the frames #4, #50, ... continue after the frame #2.

0060In Figs. 7A to 7F as well, not only the arrangement rules of the intra-frame coding and the inter-frame coding are maintained at time points before and after the insertion but also the relation of the reference frames of the inter-frame prediction is maintained, so that the reproducing operation can be simplified and the burden on the reproducing process is reduced.

0061Fig. 8 shows an example of recording track patterns for Figs. 7A to 7F.

0062A second example outside the scope of the what is claimed, in which the frames after the I frame before the insertion point is deleted for the recorded image will now be described. Fig. 9A shows a prediction relation between the inputting order (or reproducing order) of the video image signal to be insertion edited and the inter-frame coding. Fig. 9B shows a recording order (coding order). Fig. 9C shows an inputting order of the video image signal to be inserted. Fig. 9D shows a prediction relation and a coding method according to the example of the video image signal to be inserted. Fig. 9E shows a recording order (coding order) after completion of the insertion.

0063It is now assumed that the frame #50 and subsequent frames shown in Fig. 7C are inserted just before the frame #8 as an I frame. In the example , the frames for a time interval from a time point of the insertion point until a time point just before the first I frame (frame #4 in Figs. 9A to 9E) are deleted. That is, the frames #5, #6, and #7 are deleted. Further, a few first frames of the images (frame #50 and subsequent frames) to be inserted are inter-frame coded by using the past first I frame from the insertion point as a reference frame.

0064In a manner similar to the case of Figs. 5A to 5E, for the process to retrieve the recording start point of the insertion recording, the switch 84 has already been turned on and the reproduction data of the I frame (frame #4 in the example) just before the insertion point of the existing record is transferred to the movement compensation inter-frame prediction circuit 38 of the recording system and is preserved. Thus, the frame #4 can be set to the reference frame at the time of the inter-frame prediction coding of the frames #50, #51, and #52 of the images to be inserted.

0065In the case where the movement compensation inter-frame prediction circuit 38 of the recording system and the movement compensation inter-frame prediction circuit 73 of the reproducing system are commonly used, when the insertion recording is performed, it is sufficient to preserve the memory contents in the frame memories 100 and 102.

0066After completion of the insertion recording, the recording order of the coding data on the magnetic tape 54 is as shown in Fig. 9E. When the coding data is reproduced, the frame #50 and subsequent frames are reproduced after the frame #4. Namely, this equivalents to that the frame #50 and subsequent frames are joined after the frame #4.

0067In Figs. 9A to 9E, not only the arrangement rules of the intra-frame coding and the inter-frame coding are maintained at time points before and after the insertion but also the relation of the reference frames of the inter-frame prediction is maintained, so that the reproducing operation can be simplified and the burden on the reproducing process is reduced.

0068Fig. 10 shows an example of recording track patterns for Figs. 9A to 9E.

0069In the above description, the frame has been set to a coding unit. However, it is also possible to use a compression method in which the intra-field coding and the inter-field coding are commonly used by setting the field to a coding unit. The arrangement period of the intra-frame (or intra-field) coding picture plane is not limited to that shown in the above example. The arrangement and the like of the both-side prediction and one-side prediction are also not limited to those shown in the above example. Compression methods other than the combination of the DCT and the variable length coding can be also obviously applied to the invention.

0070In the case where the image data to be insertion recorded has been coded and recorded on the recording medium in a manner similar to the image data to be insertion edited, it will be obviously understood that it is preferable that the data portion which is unnecessary to change the reference frames or the like is transferred and recorded in a compressed state.

0071Although the invention has been described with respect to the case where the magnetic tape is used as a recording medium as an example, the cases of using other recording media such as magnetic disk, optical disk, magnetooptic disk, solid-state memory, and the like are also incorporated in the range of the present invention.

0072Although the above embodiment has been described with respect to the case where the invention was applied to the VTR, the transmitting apparatus of the invention can be also obviously applied to the case of inserting another channel data to certain channel data.

0073As will be obviously understood from the above description, according to the embodiment , in the transmitting apparatus using the compression method using both of the intra-frame coding and the inter-frame coding, a consistency of the reproducing operation can be maintained at time points before and after the insertion of another image. Consequently, for example, in the VTR or the like, a reproducing operation can be easily performed and the apparatus can easily correspond to the joint recording, edition, special reproduction, or the like.