An image encoding and transmitting system

Claim 2

The image encoding and transmitting system according to Claim 1, wherein said intra-frame/inter-frame determining means (22) comprises: an intra-frame/inter-frame determiner (12), said difference collating means comprises: a difference collator (31) for determining whether or not the value of one or more data in blocks involved in said second difference data is larger than a predetermined threshold, and a one-frame delay unit (11) for delaying said determination result by 1 frame when said difference collator (31) determines that one or more data in said block is larger than said threshold to compulsorily output said data in said frame to said intra-frame/inter-frame determiner (12), said image transmitting system further comprises on its sending side: a discrete cosine transform unit (13) for performing a discrete cosine transform on inputted image data, a first subtracter for obtaining the difference between said input frame data as an output of said discrete cosine transform and data obtained by multiplying the data in the preceding frame by a leak coefficient α as said constant to output the result as said first difference data to said intra-frame/inter-frame determiner (12), a quantizer (14) for quantizing an output from said intra-frame/inter-frame determiner (12), an adder (16) for outputting the sum of an output from said quantizer (14) and data obtained by multiplying the data in the preceding frame by a leak coefficient α when said intra-frame/inter-frame determiner (12) outputs inter-frame data, and outputting an output from said quantizer (14) as is when it outputs intra-frame data, a frame memory (9) for storing an output from said adder (16), a leak coefficient unit (10) for multiplying an output from said frame memory (9) by a leak coefficient α to output the result to said first subtracter (11) and said adder (16), and a second subtracter (30) for obtaining the difference between an output of said discrete cosine transform (13) and an output from said frame memory (9) to output it as said second difference data to said difference collator (31).

Claim 3

The image encoding and transmitting system according to Claim 2, wherein said difference collator (31) comprises: comparators (33) as many as the number of blocks in one frame for comparing the absolute value of said first difference data for each piece of data with a predetermined threshold, and outputting "0" when said absolute value of said first difference data is greater than said threshold and "1" when it is smaller than said threshold, and an AND circuit (34) for receiving an output of each of said comparators (33), wherein said one-frame delay unit comprises a first-in-first-out memory (35) for delaying by one frame an output from said AND circuit (34).

Claim 4

The image encoding and transmitting system according to Claim 3, wherein said intra-frame/inter-frame determiner (12) comprises: a first sum-of-squares unit (36a) for receiving an output from said first subtracter (11) and calculating a sum of squares of data in each block for said first difference data, a second sum-of-squares unit (36b) for receiving inputted frame data as an output from said discrete cosine transform unit (13), and calculating a sum of squares of data in each block for said input frame data, a comparator (37) for comparing an output from said first sum-of-squares (36a) with an output from said second sum-of-squares (36b), and outputting "0" when said output from said first sum-of-squares (36a) is larger than said output from said second sum-of-squares (36b), and "1" when it is smaller than said output from said second sum-of-squares, an AND circuit (38) for receiving an output from said comparator (37) and an output from said first-in-first-out memory (35), a first delay unit (39a) for delaying an output from said first subtracter (11) to the time point when a comparison result is outputted by said comparator (37), a second delay unit (39b) for delaying an output from said discrete cosine transform unit (13) by the delay time of said first delay unit (39a), and a selector (40) for selecting and outputting an output from said first delay unit (39a) when said AND circuit (38) outputs "1", and an output from said second delay unit (39b) when said AND circuit (38) outputs "0".

Claim 5

An image encoding and transmitting system for dividing inputted frame image data into a plurality of blocks comprising picture elements, encoding them in block units, and transmitting said encoded data in packet units, comprising on its receiving side: picture element domain frame data storing means (24) for storing received image data in the picture element domain, and motion vector detecting means (25) for detecting a motion vector using the data in the blocks surrounding the block whose encoded data were not received due to packet-discard among the frame data stored in the picture element domain frame data storage (24) when a packet-discard detection signal is inputted, and replacing using said motion vector the data in the blocks whose encoded data are not received.

Claim 6

The image encoding and transmitting system according to Claim 5, wherein a sequence number is assigned to said packet on the sending side to transmit said packet, and packet-discard is detected on the receiving side according to said sequence numbers.

Claim 7

The image encoding and transmitting system according to Claim 5, said image domain frame data storage (24) comprises a frame memory (41), said motion vector detecting means (25) comprises a motion compensation MC replacement unit (42), said image encoding and transmitting system further comprises on its receiving side: an inverse discrete cosine transform unit (17) for performing an inverse discrete cosine transform on received data, an adder (20) for obtaining the sum of the data in the present frame as an output from said inverse discrete cosine transform unit (17) and the data obtained by multiplying the data in the preceding frame by a leak coefficient α to output it to said frame memory (41), and a leak coefficient unit (19) for multiplying an output from said frame memory by a leak coefficient # to output the result to said adder (20).

Claim 8

The image encoding and transmitting system according to Claim 7, further comprising before said inverse discrete cosine transform unit (17): a header remover (103) for removing a header from a received packet, a separator (104) for separating a packet whose header is removed into a sequence number and encoded data, a counter (105) to be stepped up when said packet is received, a comparator (106) for comparing a count value outputted from said counter (105) with a sequence number outputted by said separator (104), and outputting said detection signal when the discard of a packet is detected, a memory (107) for storing encoded data outputted by said separator (104), and a read, controller (108) for outputting the content stored in said memory (107) when said comparator (106) does not detect the discard of a packet, and outputting dummy data "0" as encoded data to said inverse discrete cosine transform unit (17).

Claim 9

The image encoding and transmitting system according to Claim 7, wherein said motion compensation replacement unit (42) comprises: a 2-block-line delay unit (110) and a 1-block-line delay unit (111) for delaying said packet discard detection signal by 2 block lines and 1 block line, a B(i,j-8) discard detector (112) for detecting the discard of a block above the target block to be replaced according to an output from said 2-block-line delay unit (110), a B(i,j) discard detector (113) for detecting the discard of said target block to be replaced according to an output from said 1-block-line delay unit (111), a B(i,j+8) discard detector (114) for detecting the discard of a block below said target block to be replaced according to an input from said packet discard detection signal, a motion vector detector (115) for detecting a motion vector using the outputs from said three discard detectors (112, 113, 114) and the content stored in said frame memory (41), and a replacement controller (116) for replacing data using said detected motion vector.

Claim 10

An image encoding and transmitting system according to Claim 5, further comprising on its receiving side: coefficient domain frame data storage means for storing image data in the frequency coefficient domain, wherein the block data in the preceding frame stored in said coefficient domain frame data storing means (24) are transformed to the picture element domain, and said transformed data are stored in said picture element domain frame data storage means.

Claim 11

The image encoding and transmitting system according to Claim 10, said coefficient domain claim data storage comprises a first frame memory (18), said picture domain frame data storage (24) comprises a second frame memory (43), said motion vector detecting means (25) comprises a motion compensation (MC) replacement unit (44), said image encoding and transmitting system further comprising on its receiving side: an adder (20) for obtaining the sum of the received data in the present frame and the data obtained by multiplying the data in the preceding frame by a leak coefficient α when packet discard is not detected, and outputting said data in the preceding frame as is to said first frame memory (18) when packet discard is detected, a leak coefficient unit (19) for outputting to said adder (20) a value obtained by multiplying the output from said first frame memory (18) by a leak coefficient α when packet discard is not detected, or outputting an output from said first frame memory (18) as is to said adder (20) when packet discard is detected, and an inverse discrete cosine transform unit (17) for performing an inverse discrete cosine transform on an output from said adder (20) and output the result to said second frame memory (43).

Claim 12

The image encoding and transmitting system according to Claim 11, further comprising before said adder (20): a header remover (103) for removing a header from a received packet, a separator (104) for separating a packet whose header is removed into a sequence number and encoded data, a counter (105) to be stepped up when said packet is received, a comparator (106) for comparing a count value outputted from said counter (105) with a sequence number outputted by said separator (104), and outputting said detection signal when the discard of a packet is detected, a memory (107) for storing encoded data outputted by said separator (104), and a read controller (108) for outputting the content stored in said (107) memory when said comparator (106) does not detect the discard of a packet, and outputting dummy data "0" as encoded data to said adder (20).

Claim 13

The image encoding and transmitting system according to Claim 11, wherein said motion compensation replacement unit (44) comprises: a 2-block-line delay unit (110) and a 1-block-line delay unit (111) for delaying said packet discard detection signal by 2 block lines and 1 block line, a B(i,j-8) discard detector (112) for detecting the discard of a block above the target block to be replaced according to an output from said 2-block-line delay unit (110), a B(i,j) discard detector (113) for detecting the discard of said target block to be replaced according to an output from said 1-block-line delay unit (111), a B(i,j+8) discard detector (14) for detecting the discard of a block below said target block to be replaced according to an input from said packet discard detection signal, a motion vector detector (115) for detecting a motion vector using the outputs from said three discard detectors (112, 113, 114) and the content stored in said frame memory (41), and a replacement controller (116) for replacing data using said detected motion vector.

Claim 14

The image encoding and transmitting system according to Claim 5, wherein in a data replacement process using said motion vector, determination is made as to whether or not block B(i,j) having its upper left corner point (i,j) is discarded, if no, the process flow is terminated, if yes, the variable y indicating the vertical coordinate is set to "0", the variable x indicating the horizontal coordinate is set to "0", the horizontal coordinate (i+x) and the vertical coordinate (j+y) in the present frame are replaced with the horizontal coordinate (i+x+ x component of the motion vector) and the vertical coordinate (j+y+ y component of the motion vector) in the preceding frame, the value of the variable x indicating the horizontal coordinate is stepped up, determination is made as to whether or not "x" has reached 8, if "x" has not reached 8, the processes are repeated from the process in which said present frame data are replaced with the preceding frame data, if "x" has reached 8, the value of the variable y is stepped up, determination is made as to whether or not "y" has reached 8, if "y" has not reached 8, the processes are repeated from the process in which the variable x indicating said horizontal coordinate is set to "0", and if "y" has reached 8, the process flow is terminated.

Claim 15

The image encoding and transmitting system according to Claim 14, wherein an average value of motion vectors of one or more blocks not affected by packet-discard among blocks surrounding the block whose encoded data are not received is used as a motion vector in replacing said block data.

Claim 16

The image encoding and transmitting system according to Claim 15, wherein in a moving vector average value calculating process, determination is made as to whether or not block B(i,j) having its upper left corner point (i,j) is discarded, if no, the process flow is terminated, if yes, the variable blkct indicating the number of significant blocks for use in detecting a motion vector is set to "0", determination is made as to whether or not block B(i,j-8) above said discarded block B(i,j) is discarded, if said block B(i,j-8) is not discarded, the vector 1x as the x component vector and the vector 1y as the y component of said motion vector for said block B(i,j-8) are detected, the value of the variable blkct as said number of significant blocks is stepped up, said vector 1x is set as a vector x as the x component of a motion vector to be processed in said average value calculation, and said vector 1y is set as a vector y as the y component of a motion vector to be processed in said average value calculation, if said block B(i,j-8) is discarded, no actions are taken, determination is made whether of not block B(i,j+8) below said discarded block B(i,j) is discarded, if said block B(i,j+8) is not discarded, a vector 2x as the x component of the motion vector for said block B(i,j+8) and a motion vector 2y as the y component are detected, a value of the variable blkct indicating said number of significant blocks is stepped up, said vector 2x is added to the x component of the motion vector to be processed in said average value calculation to obtain the x component of a new motion vector, said vector 2y is added to the y component of the motion vector to be processed in said average value calculation to obtain the y component of a new motion vector, if said block B(i,j+8) is discarded, no actions are taken, determination is made as to whether or not said variable blkct equals "0", if said variable blkct is "0", the x and y components of the motion vector are set to "0" and the process flow is terminated, and if said variable blkct is not "0", the x component of the motion vector to be processed in said average value calculation is divided by the variable blkct, and the result is assigned as the x component of the motion vector, the y component of the motion vector to be processed in said average value calculation is divided by the variable blkct, and the result is assigned as the y component of the motion vector, and the process flow is terminated.

Claim 17

The image encoding and transmitting system according to Claim 16, wherein in a motion vector detecting process for said undiscarded block, a large value is set as the initial value of sum 1 which is a sum of absolute values, for all data in a block, of difference data between received data in a block whose motion vector is to be detected in the present frame and its corresponding data in the preceding frame, the initial value of the variable vecy indicating the y component of a motion vector is set to "-4", the initial value of the variable vecx indicating the x component of a motion vector is set to "-4", the sum "sum 2" of the absolute values of the differences between each piece of the present frame data for the point (i+x, j+y) and the preceding frame data for the point (i+x+vecx, j+y+vecy) for all blocks is obtained, determination is made as to whether or not the value of said sum 2 is smaller than the value of said sum 1, if yes, vecx is set as the x component of a motion vector, vecy as its y component, and the value of sum 2 is substituted for sum 1, if no, no actions are taken, the value of the variable vecx is stepped up, determination is made as to whether or not the value of vecx has reached "5", if not, the processes are repeated from the process in which said sum 2 is obtained, if yes, the value of the variable vecy is stepped up, determination is made as to whether of not the value of vecy has reached "5", if not, the processes are repeated from the process in which the value of said variable vecx is set to the initial value -4, if yes, the process flow is terminated.

Claim 18

The image encoding and transmitting system according to Claim 14, wherein a motion vector is detected in a block group which contains one or more blocks not affected by packet discard among blocks surrounding a block whose encoded data are not received, and said motion vector is used for replacing said block data.

Claim 19

The image encoding and transmitting system according to Claim 18, wherein in a motion vector calculating process for said block group, determination is made as to whether or not block B(i,j) having its upper left corner point (i,j) is discarded, if no, the process flow is terminated, if yes, determination is made as to whether or not block B(i,j-8) above said block B(i,j) is discarded, if yes, the minimum value mini of the vertical coordinate of said group processed in calculating said vector is set to "j", if not, said value minj is set to "j-8", determination is made as to whether or not block B(i,j+8) below said discarded block B(i,j), if yes, the maximum value maxj of the vertical coordinate of said group processed in calculating said vector is set to "j+7", if not said value maxj is set to "j+15", determination is made as to whether or not said value "minj" equals "j" and said value "maxj" equals "j+7", if yes, the x and y components of said motion vector are set to "0", and the process flow is terminated, and if not, a motion vector is detected in a block group which does not contain a discarded block between said minimum value "minj" and said maximum value "maxj" of the vertical coordinate, and the process flow is terminated.

Claim 20

The image encoding and transmitting system according to Claim 19, wherein in a motion vector detecting process for said undiscarded block, a large value is set as the initial value of sum 1 which is the minimum sum of absolute values of difference data between received data in a block whose motion vector is to be detected in the present frame and its corresponding data in the preceding frame, the initial value of the variable vecy indicating the y component of a motion vector is set to "-4", the initial value of the variable vecx indicating the x component of a motion vector is set to "-4", the sum "sum 2" of the absolute values of the differences between each piece of the present frame data for the point (i+x, j+y) and the preceding frame data for the point (i+x+vecx, j+y+vecy) for all blocks in said group is obtained, determination is made as to whether or not the value of said sum 2 is smaller than the value of said sum 1, if yes, vecx is set as the x component of a motion vector, vecy as its y component, and the value of sum 2 is substituted for sum 1, if no, no actions are taken, the value of the variable vecx is stepped up, determination is made as to whether or not the value of vecx has reached "5", if not, the processes are repeated from the process in which said sum 2 is obtained, if yes, the value of the variable vecy is stepped up, determination is made as to whether of not the value of vecy has reached "5", if not, the processes are repeated from the process in which the value of said variable vecx is set to the initial value -4, if yes, the process flow is terminated.

Claim 21

An image encoding and transmitting system used in an image transmission system for dividing inputted frame image data into blocks comprising a plurality of picture elements, encoding said data in block units, and transmitting said encoded data in packet units, said system comprising: a motion vector detector (100) on a sending side of said image transmission system for detecting a motion vector for each of said block and sending said motion vector to a receiving side, an inverse discrete cosine transform unit (17) on said receiving side for performing an inverse discrete cosine transform on received data, an adder (20) for adding the present frame data as an output from said discrete cosine transform unit (17) to data obtained by multiplying the delay result obtained, by delaying data in the preceding frame according to a motion vector received from said sending side by a leak coefficient α, a frame memory (18) for storing an output of said adder (20), a motion compensation replacement unit (MC)(118) for replacing the data for a block whose encoded data are not received due to packet-discard using a motion vector received from said sending side and the content shored in said frame memory (18) when a packet discard detection signal is inputted, a variable delay unit (102) for delaying an output from said frame memory (18) according to a motion vector received from said sending side, a leak coefficient unit (19) for multiplying an output from said variable delay unit (102) by a leak coefficient α and outputting the product to said adder (20).

Claim 22

The image encoding and transmitting system according to Claim 21, further comprising before said inverse discrete cosine transform unit (17): a header remover (103) for removing a header from a received packet, a separator (104) for separating a packet whose header is removed into a sequence number and encoded data, a counter (105) to be stepped up when said packet is received, a comparator (106) for comparing a count value outputted from said counter (105) with a sequence number outputted by said separator (104), and outputting said detection signal when the discard of a packet is detected, a memory (107) for storing encoded data outputted by said separator (104), and a read controller (108) for outputting the content stored in said memory (107) when said comparator (106) does not detect the discard of a packet, and outputting dummy data "0" as encoded data to said inverse discrete cosine transform unit (17).

Claim 23

The image encoding and transmitting system according to Claim 21, wherein said motion compensation replacement unit (118) comprises: a 2-block-line delay unit (110) and a 1-block-line delay unit (111) for delaying said packet discard detection signal by 2 block lines and 1 block line, a B(i,j-8) discard detector (112) for detecting the discard of a block above the target block to be replaced according to an output from said 2-block-line delay unit (10), a B(i,j) discard detector (113) for detecting the discard of said target block to be replaced according to an output from said 1-block-line delay unit (111), a B(i,j+8) discard detector (114) for detecting the discard of a block below said target block to be replaced according to an input from said cell-discard detection signal, 1-block-line and 2-block-line delay units (120, 121) for delaying a motion vector received from said sending side by 1 block and 2 blocks respectively, a motion vector detector (122) for detecting a motion vector using the outputs from said three discard detectors (112, 113, 114) and two delay units (120, 121), and the present value of a motion vector received from said sending side, and a replacement controller (116) for replacing data using said detected motion vector.

Claim 24

The image encoding and transmitting system according to Claim 21, wherein an average value of motion vectors of one or more blocks not affected by packet-discard among blocks surrounding the block whose encoded data are not received is used as a motion vector in replacing said block data.

Claim 25

The image encoding and transmitting system according to Claim 24, wherein in a moving vector average value calculating process, determination is made as to whether or not block B(i,j) having its upper left corner point (i,j) is discarded, if no, the process flow is terminated, if yes, the variable blkct indicating the number of significant blocks for use in detecting a motion vector is set to "0", determination is made as to whether or not block B(i,j-8) above said discarded block B(i,j) is discarded, if said block B(i,j-8) is not discarded, the x component of said motion vector for said block B(i,j-8) received from the transmitting side is set to vector 1x, and the y component of said motion vector is set to vector 1y, the value of the variable blkct as said number of significant blocks is stepped up, said vector 1x is set as a vector x as the x component of a motion vector to be processed in said average value calculation, and said vector 1y is set as a vector y as the y component of a motion vector to be processed in said average value calculation, if said block B(i,j-8) is discarded, no actions are taken, determination is made whether of not block B(i,j+8) below said discarded block B(i,j) is discarded, if said block B(i,j+8) is not discarded, the x component of the motion vector for said block B(i,j+8) received from the sending side is set to vector 2x, and a the y component of the motion vector is set to vector 2y, a value of the variable blkct indicating said number of significant blocks is stepped up, said vector 2x is added to the x component of the motion vector to be processed in said average value calculation to obtain the x component of a new motion vector, said vector 2y is added to the y component of the motion vector to be processed in said average value calculation to obtain the y component of a new motion vector, if said block B(i,j+8) is discarded, no actions are taken, determination is made as to whether or not said variable blkct equals "0", if said variable blkct is "0", the x and y components of the motion vector are set to "0" and the process flow is terminated, and if said variable blkct is not "0", the x component of the motion vector to be processed in said average value calculation is divided by the variable blkct, and the result is assigned as the x component of the motion vector, the y component of the motion vector to be processed in said average value calculation is divided by the variable blkct, and the result is assigned as the y component of the motion vector, and the process flow is terminated.