| 摘要 |
<p>PROBLEM TO BE SOLVED: To efficiently compress an image with high resolution and to compress/expand an image without any distortion. SOLUTION: The image compressor 30 is provided with a predictive value generating section 31, a differential value generating section 32, a Hadamard transformation section 33, a quantization section 34, a bit separate section 35, a zigzag scan section 36, and a Huffman coding section 37. The prediction value generating section 31 generates a prediction value based on pixel data configuring received image data. The differential value generating section 32 generates a difference by taking a difference between the predicted value and pixel data. The Hadamard transformation section 33 applies Hadamard transformation the difference to obtain a Hadamard transformation coefficient. The quantization section 34 quantizes the Hadamard transformation coefficient to obtain a quantized Hadamard transformation coefficient. The bit separation section 35 applies bit separation to the quantized Hadamard transformation coefficient to obtain a high-order quantized Hadamard transformation coefficient and a low-order quantized Hadamard transformation coefficient. The zigzag scan section 36 applies zigzag scanning to the high-order quantized Hadamard transformation coefficient and the Huffman coding section 37 encodes an output of the zigzag scan section 36.</p> |
