【正文】 sed in this chapter provides high pression ratios using DCT (discrete cosine transform) – Image data divided into blocks of 8 x 8 pixels – 3 steps performed on each block ? DCT ? Quantization ? Huffman encoding 8 Embedded Systems Design: A Unified Hardware/Software Introduction, (c) 2023 Vahid/Givargis DCT step ? Transforms original 8 x 8 block into a cosinefrequency domain – Upperleft corner values represent more of the essence of the image – Lowerright corner values represent finer details ? Can reduce precision of these values and retain reasonable image quality ? FDCT (Forward DCT) formula – C(h) = if (h == 0) then 1/sqrt(2) else ? Auxiliary function used in main function F(u,v) – F(u,v) = 188。 for(rowIndex=0。 colIndexSZ_COL。 } void UartSend(char d) { fprintf(outputFileHandle, %i\n, (int)d)。 idx++。 } 24 Embedded Systems Design: A Unified Hardware/Software Introduction, (c) 2023 Vahid/Givargis CNTRL (controller) module ? Heart of the system ? CntrlInitialize for consistency with other modules only ? CntrlCaptureImage uses CCDPP module to input image and place in buffer ? CntrlCompressImage breaks the 64 x 64 buffer into 8 x 8 blocks and performs FDCT on each block using the CODEC module – Also performs quantization on each block ? CntrlSendImage transmits encoded image serially using UART module void CntrlSendImage(void) { for(i=0。 i++) for(j=0。 /* initialize the modules */ UartInitialize(uartOutputFileName)。 void UartInitialize(const char *outputFileName) { outputFileHandle = fopen(outputFileName, w)。 l++) { buffer[i * 8 + k][j * 8 + l] = CodecPopPixel()。 jSZ_COL。 return (short)(r * .25 * C(u) * C(v))。 idx++。 rowIndex = 1。 CcdCapture()。 rowIndex = 1。 if( ++rowIndex == SZ_ROW ) { colIndex = 1。 } } } rowIndex = 0。 colIndex++) { buffer[rowIndex][colIndex] = bias。 x8。 int x。 UartSend(((char*)temp)[0])。 l8。 /* simulate functionality */ CntrlCaptureImage()。 CntrlSendImage()。 CodecDoFdct()。 /* send lower byte */ } } } define SZ_ROW 64 define SZ_COL 64 define NUM_ROW_BLOCKS (SZ_ROW / 8) define NUM_COL_BLOCKS (SZ_COL / 8) static short buffer[SZ_ROW][SZ_COL], i, j, k, l, temp。 x8。 y8。 colIndex = 0。 } 20 Embedded Systems Design: A Unified Hardware/Software Introduction, (c) 2023 Vahid/Givargis CCDPP (CCD PreProcessing) module ? Performs zerobias adjustment ? CcdppCapture uses CcdCapture and CcdPopPixel to obtain image ? Performs zerobias adjustment after each row read in define SZ_ROW 64 define SZ_COL 64 static char buffer[SZ_ROW][SZ_COL]。 } } return pixel。 static unsigned rowIndex, colIndex。 colIndex = 1。 if( ++colIndex == SZ_COL ) { colIndex = 0。 } void CodecPushPixel(short p) { if( idx == 64 ) idx = 0。 x8。 iSZ_ROW。 k++) for(l=0。 void UARTInitialize(void) {} void UARTSend(unsigned char d) { while( U_STAT_REG == 1 ) { /* busy wait */ } U_TX_REG = d。 CcdppInitialize()。 j++) { for(k=0。 i++) for(j=0。 } 23 Embedded Systems Design: A Unified Hardware/Software Introduction, (c) 2023 Vahid/Givargis CODEC (cont.) ? Implementing FDCT formula C(h) = if (h == 0) then 1/sqrt(2) else F(u,v) = 188。 void CodecInitialize(void) { idx = 0。 } bias = (CcdPopPixel() + CcdPopPixel()) / 2。 rowIndex++) { for(colIndex=0。 x C(u) x C(v) Σx=0..7 Σy=0..7 Dxy x cos(π(2u + 1)u/16) x cos(π(2y + 1)v/16) ? Gives encoded pixel at row u, column v ? Dxy is original pixel value at row x, column y ? IDCT (Inverse DCT) – Reverses process to obtain original block (not needed for this design) 9 Embedded Systems Design: A Unified Hardware/Software Introduction, (c) 2023 Vahid/Givargis Quantization step ? Achieve high pression ratio by reducing image quality – Reduce bit precision of encoded data ? Fewer bits needed for encoding ? One way is to divide all values by a factor of 2 – Simple right shifts can do this – Dequantization would reverse process for depression 1150 39 4 3 1 0 26 8 3 11 41 8 1 3 115 7 3 6 2 22 514 1 1 1 4 2 26 3 17 3 82 6 1 1 3 1 2 36 2 3 1 8 544 13 37 4 10 2 1 7 836 1 1 9 4 20 2 8 2 1 14 1 9 7 21 6 3 3 12 2 15 1 3 1 1 1 7 4 1 7 4144 5 5 1 3 1 0 1 5 1 0 0 14 9 1 0 3 12 1 0 5 3 0 2 50 8 2 2 5 3 2 16 2 5 1 1 3 1 15 1 1 1 3 4 3 22 1 3 1 0 0 2 31 2 1 2 1 0 1 1After being decoded using DCT After quantization Divide each cell’s value by 8 10 Embedded Systems Design: A Unified Hardware/Software Introduction, (c) 2023 Vahid/Givargis ? Serialize 8 x 8 block of pixels – Values are converted into single list using zigzag pattern ? Perform Huffman encoding – More frequently occurring pixels assigned short binary code – Longer binary codes left for less frequently occurring pixels ? Each pixel in serial list converted to Huffman encoded values – Much shorter list, thus pression Huffman encoding step 11 Embedded Systems Design: A Unified Hardware/Software Introduction, (c) 2023 Vahid/Givargis Huffman encoding example ? Pixel frequencies on left – Pixel value –1 occurs 15 times – Pixel value 14 occurs 1 time