TH16705B - Image encoder Picture decoder How to encrypt pictures How to decrypt pictures Image encryption program recording media And recording media, image decoding program - Google Patents

Abstract

DC60 (12/01/47) The purpose of this invention is to encode an effective image or shape signal. More pictures than previous invention Along the path to achieve that objective Pig detector The transformation cell 2 receives the input signal 1 as an input signal and measures pixels which Change the two pixel values. Also, the change pixel predictor 4 reads the reference image. Which is stored in memory 3, and predicts the change of the input signal only, the difference calculator 5 subtracts the output of the shift predictor 4 from the output. The measurement of the pixel transducer, the change 2, the difference 7, compares the tolerance e and the error, the prediction D and the output X, which requires the number of bits. The minimum for encoding in De value is less than or equal to x, less than or equal to D + e, our output. The difference digit 7 is encoded by decoder 8 to become the encoded signal. 9 In addition, our output, the difference modifier 7 is added in, adds a difference value 11 to the predicted pixel 4 of the predicted pixel pitch predictor 4 with the pixel method. The transformation is computed and the pixel decoder changes 10 pixels from the coded pixels displayed by the predictor. 4 transition pixels to transition pixels are decoded to be stored in memory 3. This invention aims to encode a video signal or shape signal that is more efficient than the previous invention along the way to achieve that purpose. The change-pixel detector 2 accepts the input signal 1 as the input signal and measures the two-pixel shift values. Variation Pizel Predictor 4 Read reference images. Which is stored in memory 3, and predicts a specific change in input signal. The difference calculator 5 subtracts the transformation pixel predictor output 4 from the transformation pixel detector output 2, the difference matrix 7, compare the tolerance e. And error, predictions D and output X, which requires the least amount of torque for encoding in De (formula) x (formula) D + e, the difference value routers output. 7 is encoded by decoder 8 to become the encoded signal 9 In addition to our output, the differential 7 is added in the differential 11, solving the predicted pixel 47. Of a predicted pixel pitch 4 with the way the change pixels are calculated And various 10-pixel transition pixels from a coded pixel lighted by a 4 transition pixel to a transition pixel is decoded to be stored in memory3.

Claims (3)

1. Image decoding device for receiving and decoding the encoded signal. It includes: One decoding method for decoding the encoded signal to obtain the differential signal differential, and for the differential output obtained as the pixel difference. Deciphered; Calculation path for incremental, decoded pixel value and compensated pixel value signal, and for output, the result of increase as decoded pixel value signal; A second decoding method for decoding the encoded signal to obtain the vectors. Signal motion, pixel value, and motion vector of shape signal. And for the output of these motion vectors as motion vectors, decoded pixel values and decoded shape motion vectors respectively; One of the trajectory compensation for the reference image's pixel value compensation using decoded pixel-value motion vector. And for the output the result of displacement compensation as a signal, the pixel value is adjusted chuck; A second motion compensation path for image contour compensation. Reference that uses the decoded shape motion vector. And for the output the result of displacement compensation as a compensated shape signal; And a third decoding path for decoding the input encoded signal according to the compensated shape signal. And for signal output that is decoded as a shape signal 2. Image decoding device according to claim 1, where when the input encoded signal includes the difference information which is obtained by encoding the difference between the motion vector of the previously encoded shape signal. With the shape signal motion vector, which is measured from the input encoded signal. The second decoding path is also available for decoding the difference information and for adding the decoded difference to the previously decoded shape motion vector to obtain the decoded shape motion vector. 3. Image decoding device according to claim 1, where when the input encoded signal includes the difference information obtained by encoding the difference between the signal motion vector, the pre-encoded pixel value. Page to the signal motion vector, the pixel value which is measured from the input signal. The second decoding path is also available for decoding the difference information and for adding the decoded difference value to the previously decoded pixel-value signal motion vector to obtain the decoded pixel-value motion vector.