Image processing device and method

Abstract

An image processing device and method capable of improving image quality of a decoded image and of an image to be referenced from now on with motion compensation. A mosquito noise filter is provided within a motion compensation loop including at least a computing unit, an orthogonal transform unit, a quantization unit, an inverse quantization unit, an inverse orthogonal transform unit, a computing unit, a frame memory, a switch, a motion prediction/compensation unit, and a prediction image selecting unit. The mosquito noise filter uses information from the orthogonal transform unit, quantization unit, and a lossless encoding unit to determine whether to perform filter processing for removing mosquito noise. The device and method may be applied to an image encoding device for performing encoding with H.264/AVC format, for example.

1 - 14 . (canceled) 
     
     
         15 . An image processing device comprising:
 filter processing means configured to perform, within a motion compensation loop of an image, filter processing for removing mosquito noise on a block to be processed in said image under control according to an orthogonal transform size of said block and parameters obtained from generated code amount and quantization scale regarding said block; and   encoding means configured to encode said image and said encoding information.   
     
     
         16 . The image processing device according to  claim 15 , wherein said filter processing means includes:
 threshold determining means configured to determine a threshold according to the orthogonal transform size of said block;   difficulty level parameter calculating means configured to calculate a difficulty level parameter of said block using information relating to said generated code amount and a quantization scale as to said block; and   filter processing control means configured to perform control so as to perform said filter processing on said block in the case that said difficulty level parameter calculated by said difficulty level parameter calculating means is greater than said threshold determined by said threshold determining means.   
     
     
         17 . The image processing device according to  claim 16 , wherein said threshold includes an offset value that a user can set. 
     
     
         18 . The image processing device according to  claim 15 , wherein said filter processing means uses generation bits as information relating to said generated code amount. 
     
     
         19 . The image processing device according to  claim 15 , wherein said filter processing means uses a generation bit or generation bin as information relating to said generated code amount in the case that CABAC is used as a lossless encoding format. 
     
     
         20 . The image processing device according to  claim 15 , wherein syntax elements in said encoding information include a quantization parameter in said block, and a number of non-zero orthogonal transform coefficients after quantization. 
     
     
         21 . The image processing device according to  claim 20 , wherein said filter processing means includes:
 threshold determining means configured to determine a threshold according to the orthogonal transform size and quantization parameter of said block; and   filter processing control means configured to perform control so as to perform said filter processing on said block in the case that said number of non-zero orthogonal transform coefficients after quantization is greater than said threshold determined by said threshold determining means.   
     
     
         22 . The image processing device according to  claim 20 , wherein said threshold includes an offset value that a user can set. 
     
     
         23 . The image processing device according to  claim 15 , wherein syntax elements in said encoding information include motion vector information as to said block. 
     
     
         24 . The image processing device according to  claim 15 , wherein said filter processing means performs smoothing processing using a two-dimensional filter having a predetermined window size with a pixel to be processed as a center regarding a pixel value included in said block as said filter processing. 
     
     
         25 . An image processing method comprising:
 performing, with filter processing means of an image processing device, within a motion compensation loop of an image, filter processing for removing mosquito noise on a block to be processed in said image under control according to an orthogonal transform size and syntax elements in encoding information of said block; and   encoding, with encoding means, said image and said encoding information.   
     
     
         26 . An image processing device comprising:
 filter processing means configured to perform, within a motion compensation loop of an image, filter processing for removing mosquito noise on a block to be processed in said image under control according to an orthogonal transform size and syntax elements in encoding information of said block; and   decoding means configured to decode said image and said encoding information that have been encoded.   
     
     
         27 . An image processing method comprising:
 performing, with filter processing means of an image processing device, within a motion compensation loop of an image, filter processing for removing mosquito noise on a block to be processed in said image under control according to an orthogonal transform size and syntax elements in encoding information of said block; and   decoding, with decoding means, said image and said encoding information that have been encoded.