IMAGE ENCODING METHOD, IMAGE DECODING METHOD, IMAGE ENCODING DEVICE AND IMAGE DECODING DEVICE
First Claim
1. An image encoding method for receiving pixel data of a pixel to be compressed, and compressing the input pixel data, comprising:
- a predicted pixel generation step of generating a predicted value for the pixel data based on at least one of pixels adjacent to the pixel to be compressed;
a quantization step of quantizing a difference value between the pixel data and the predicted value generated in the predicted pixel generation step to a quantized value having a smaller number of bits than that of the pixel data, to compress the pixel data to the quantized value; and
a quantization width determination step of determining a quantization width used in the quantization step,whereinin the quantization width determination step, it is determined whether or not there are both a flat portion and an edge portion, in groups of a plurality of the pixels, and based on a result of the determination, quantization widths used in the quantization step are adaptively changed on a pixel-by-pixel basis.
4 Assignments
0 Petitions
Accused Products
Abstract
In an image encoding device, a predicted pixel generator generates a predicted value for pixel data based on at least one of pixels adjacent to a pixel to be compressed. A quantization processor quantizes a difference value between the pixel data and the generated predicted value to a quantized value having a smaller number of bits than that of the pixel data to compress the pixel data to the quantized value. In a quantization width determiner, an edge determiner determines whether or not there are both a flat portion and an edge portion, based on a characteristic of the difference value in the group, and an edge pixel determiner determines whether each of pixels included in the group is a pixel in an edge portion or a pixel in a flat portion, based on the difference value, whereby a quantization width used in the quantization processor is determined.
28 Citations
27 Claims
-
1. An image encoding method for receiving pixel data of a pixel to be compressed, and compressing the input pixel data, comprising:
-
a predicted pixel generation step of generating a predicted value for the pixel data based on at least one of pixels adjacent to the pixel to be compressed; a quantization step of quantizing a difference value between the pixel data and the predicted value generated in the predicted pixel generation step to a quantized value having a smaller number of bits than that of the pixel data, to compress the pixel data to the quantized value; and a quantization width determination step of determining a quantization width used in the quantization step, wherein in the quantization width determination step, it is determined whether or not there are both a flat portion and an edge portion, in groups of a plurality of the pixels, and based on a result of the determination, quantization widths used in the quantization step are adaptively changed on a pixel-by-pixel basis. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 27)
-
-
17. An image decoding method, wherein
input encoded data contains a quantized value, and the quantized value is obtained by quantizing a difference value between pixel data and a predicted value for the pixel data, and the predicted value for the pixel data is generated based on at least one of pixels adjacent to a pixel of the pixel data, and the quantized value is decoded to the pixel data, and the method comprises: -
a predicted pixel generation step of generating the predicted value for the current pixel data based on the past pixel data which has already been decoded; an inverse quantization step of inverse quantizing the quantized value to generate the difference value having the quantized value which has been inverse quantized; and a quantization width determination step of determining a quantization width used in the inverse quantization step, on a pixel-by-pixel basis, based on encoded quantization information and an encoded quantized value, and in the quantization width determination step, quantization widths used in the inverse quantization step are adaptively changed on a pixel-by-pixel basis.
-
-
18. An image encoding device for receiving pixel data of a pixel to be compressed, and compressing the input pixel data, comprising:
-
a predicted pixel generator configured to generate a predicted value for the pixel data based on at least one of pixels adjacent to the pixel to be compressed; a quantizer configured to quantize a difference value between the pixel data and the predicted value generated by the predicted pixel generator to a quantized value having a smaller number of bits than that of the pixel data to compress the pixel data to the quantized value; and a quantization width determiner configured to determine a quantization width used in the quantizer, wherein the quantization width determiner determines whether or not there are both a flat portion and an edge portion, in groups of a plurality of the pixels, and based on a result of the determination, adaptively changes quantization widths used in the quantizer on a pixel-by-pixel basis. - View Dependent Claims (20, 21, 22, 23, 24, 25, 26)
a predicted pixel generator configured to generate the predicted value for the current pixel data based on the past pixel data which has already been decoded; an inverse quantizer configured to inverse quantize the quantized value to generate the difference value having the quantized value which has been inverse quantized; and a quantization width determiner configured to determine a quantization width used in the inverse quantizer, on a pixel-by-pixel basis, based on encoded quantization information and an encoded quantized value, and the quantization width determiner adaptively changes quantization widths used in the inverse quantizer on a pixel-by-pixel basis.
-
-
24. A digital camcorder comprising the image encoding device of claim 18, and comprising an image decoding device, wherein
input encoded data contains a quantized value, and the quantized value is obtained by quantizing a difference value between pixel data and a predicted value for the pixel data, and the predicted value for the pixel data is generated based on at least one of pixels adjacent to a pixel of the pixel data, and the quantized value is decoded to the pixel data, and the image decoding device comprises: -
a predicted pixel generator configured to generate the predicted value for the current pixel data based on the past pixel data which has already been decoded; an inverse quantizer configured to inverse quantize the quantized value to generate the difference value having the quantized value which has been inverse quantized; and a quantization width determiner configured to determine a quantization width used in the inverse quantizer, on a pixel-by-pixel basis, based on encoded quantization information and an encoded quantized value, and the quantization width determiner adaptively changes quantization widths used in the inverse quantizer on a pixel-by-pixel basis.
-
-
25. A surveillance camera comprising the image encoding device of claim 18, and comprising an image decoding device, wherein
input encoded data contains a quantized value, and the quantized value is obtained by quantizing a difference value between pixel data and a predicted value for the pixel data, and the predicted value for the pixel data is generated based on at least one of pixels adjacent to a pixel of the pixel data, and the quantized value is decoded to the pixel data, and the image decoding device comprises: -
a predicted pixel generator configured to generate the predicted value for the current pixel data based on the past pixel data which has already been decoded; an inverse quantizer configured to inverse quantize the quantized value to generate the difference value having the quantized value which has been inverse quantized; and a quantization width determiner configured to determine a quantization width used in the inverse quantizer, on a pixel-by-pixel basis, based on encoded quantization information and an encoded quantized value, and the quantization width determiner adaptively changes quantization widths used in the inverse quantizer on a pixel-by-pixel basis.
-
-
26. A solid imaging element comprising:
the image encoding device of claim 18.
-
19. An image decoding device, wherein
input encoded data contains a quantized value, and the quantized value is obtained by quantizing a difference value between pixel data and a predicted value for the pixel data, and the predicted value for the pixel data is generated based on at least one of pixels adjacent to a pixel of the pixel data, and the quantized value is decoded to the pixel data, and the device comprises: -
a predicted pixel generator configured to generate the predicted value for the current pixel data based on the past pixel data which has already been decoded; an inverse quantizer configured to inverse quantize the quantized value to generate the difference value having the quantized value which has been inverse quantized; and a quantization width determiner configured to determine a quantization width used in the inverse quantizer, on a pixel-by-pixel basis, based on encoded quantization information and an encoded quantized value, and the quantization width determiner adaptively changes quantization widths used in the inverse quantizer on a pixel-by-pixel basis.
-
Specification