University:
Boston University
Course:
CAS CS 501 | Computer Science Practicum
Academic year:

2024
Views:

323

Pages:

19
Author:

Bresolini7f1a

Introduction to ImageJ Tutorial Outline  Image Format Types  Vector Images  Raster Images  Raster Image Quality  Color  Digital vs. Print  RGB and CMYK  Getting Started with ImageJ  Opening and saving images  Selections and cropping  Image Adjustments  Measurements      Area and dimensions Histograms ROI and masking Thresholding Quantification of intensity  Macros  Recording and running  Writing your own Image File Formats  Raster  TIFF, PNG, BMP, JPEG, GIF, JPEG200 (and many more!)  i.e. files ending in .tif, .png, .bmp, .jpg, .gif, .jp2  Vector         Portable Document Format - .pdf PowerPoint - .ppt Adobe Illustrator - .ai Photoshop - .psd PostScript - .ps Encapsulated PostScript - .eps Drawing and CAD packages Scalable Vector Graphics - .svg Vector vs. Raster  Vector images can be scaled to any size or resolution.  Typically produced when creating computer graphics (plotting, drawing, diagrams, text, etc.)  Example: TrueType fonts  Raster images:  Scaling up or down requires a mathematical transform on the original image.  Re-scaled raster images are not identical to the original, and the original image may or not be recoverable from a re-scaled version.  Produced by physical image acquisition: cameras, scanners, etc. Vector Image Files  Many packages create their own proprietary formats.  Vector files are usually resolution independent.  Thus, they can be easily printed at different resolutions.  Line thickness may be an issue.  Converting to raster image format should be done with care and only as a last step when necessary.  The transformation to a raster image is completely nonreversible.  Use anti-aliasing if possible when doing the conversion. Vector Image Files – PostScript Example PostScript code %! 306 427 moveto 0 setlinewidth 240 330 lineto /Times-Roman 323 332 lineto findfont 306 427 lineto 12 scalefont setfont 306 427 moveto 0.0 setgray 323 332 lineto 0.000 setgray 374 328 lineto newpath 306 427 lineto 0 0 moveto 306 427 moveto 374 328 moveto 374 328 lineto 323 332 lineto 287 326 lineto 240 330 lineto 306 427 lineto 287 326 lineto 0 setlinewidth 374 328 lineto stroke 306 427 moveto showpage 287 326 lineto 240 330 lineto 306 427 lineto   Vector images store image information as specifications for a set of polygons. To display, the polygons are mathematically scaled to the desired size and resolution and then rendered (rasterized, i.e. converted to a raster image). Raster Image Files  Raster image files are composed of a grid of pixels.  Each pixel contains color or intensity information  File formats support many types of raster image      24 bit full color – RGB (8 bits per color) – png, tiff, rgb, bmp, jpg, jp2 32 bit full color with alpha – png, tiff, rgb, bmp, jpg, jp2 8 bit color map – gif, tiff, rgb Bitmap (black and white only) – tiff, bmp. Grayscale – can be 8,16, or 24-bit – png, tiff, bmp, jpg, jp2 Bits # of bits per pixel (N) Range of intensity levels (0  2N-1) Image Type Computer Representation (C types) 8 0 - 255 grayscale unsigned char (8 bits) 10 0 - 1023 grayscale unsigned short (16 bits) 12 0 - 4095 grayscale unsigned short (16 bits) 16 0 - 65,535 grayscale unsigned short (16 bits) 24 0 – 16,777,215 grayscale unsigned int (32 bits) 3x8 0 - 255 per channel RGB unsigned int (32 bits) 4x8 0 - 255 per channel RGB with alpha unsigned int (32 bits)  The number of bits determines the number of intensity levels that can be digitally encoded.  For non-power-of-two bits per pixel the image is usually stored in the next higher-up pixel format.  Example: 10-bit image stored as a 16-bit TIFF. Only values 0-1023 would be used. What sets the # of bits and pixels?  The physical device capturing the image sets the number of pixels  The electronics that convert light intensity to an electronic signal sets the number of bits. The image as projected onto the CCD chip. Pixel (spatial) resolution set when chip is manufactured. Charge accumulates per pixel during exposure. Charge levels are read out by a digitizer with a fixed number of bits per pixel. Charge levels over the maximum bit value are saturated. 232 172 234 255 209 119 198 222 187 155 162 188 206 178 195 245 8-bit digitization assigns the values shown. Color Acquisition  Color in an image can be acquired in many ways. Here’s 2 out of many.  An RGB CCD or CMOS chip has 4 sub-pixels per pixel of spatial resolution that detect red, green, or blue  A filter wheel is used to select a range of wavelengths, and multiple images are taken with a monochromatic chip.  RGB images use 3 separate acquisitions  “R” “G” “B” bands are user-selectable.  More than 3 bands can be used for a multispectral image. Factors Involving Raster Image Quality  DPI: Dots per Inch  A term for printing where dots of ink are physical printed to a page  PPI: Pixels per Inch is the digital equivalent 214   Resolution: the number of pixels in the image (width x height)  The number of pixels in an inch (as on a monitor)  DPI and PPI are often used interchangeably  195   Number of available colors in the image format.  Anti-aliasing: Essentially a blurring operation that is done to smooth images and graphics when they are re-sized.  Compression 850 x 1134, RGB Compression and Image Quality  Compression in image formats saves size on disk  Compressed images take longer to open and save  Lossless compression:  PNG format, TIFF, JPEG2000, others  Does not alter the image  A ZIP file is a non-image example of lossless compression  Lossy compression:      JPEG format, GIF, JPEG2000 (careful!), others Discards information from the image Much smaller file sizes than lossless compression Greater the compression, lower the quality Not suitable for the storage of scientific images!! Format Size (bytes) TIFF 2,891,854 PNG 2,630,062 JPEG (77% quality) 220,517 JPEG (95% quality) 449,654 JPEG Compression JPEG Quality: 95% 50% Original Orig – JPEG: Same B&C)   The JPEG algorithm compresses an image by removing detail from the image before performing a lossless compression High-frequency information is discarded   Sharp transitions in contrast Changes in color hue 15% Color Image File: 24 Bit Full Color Pixel Values (28x28x28=16.7 million possible colors) R G 0 24 bit 3x3 Pixel P3 PPM File 33 255 0 0 0 128 0 0 255 0 0 128 128 128 0 128 0 0 255 0 255 255 255 0 0 128 0 0 255 B 0 R 0 G 128 B R 0 0 128 128 128 0 128 0 255 255 255 0 0 128 G 255 B 0 0 0 255 0 0 0 255 Web Equivalents (Hexadecimal RRGGBB) 000000 800000 FF0000 808080 008000 00FF00 FFFFFF 000080 0000FF Color Image File: 8 Bit Color Map - aka Indexed (256 possible colors) Color name Index # R G B Black 0 0 0 0 Red 1 255 0 0 Yellow 2 255 255 0 Green 3 0 255 0 Cyan 4 0 255 255 Blue 5 0 0 255 Magenta 6 255 0 255 Dk red 7 128 0 0 Dk green 8 0 128 0 Dk blue 9 0 0 128 . Gray 250 128 128 128 . White 255 255 255 255 0 250 255 7 8 9 1 3 5 An 8-bit value per pixel becomes an index in the 256-element lookup table, the color map. So Far…  Image Format Types  Vector Images  Raster Images  Raster Image Quality  Color  Digital vs. Print  RGB and CMYK  Getting Started with ImageJ  Opening and saving images  Selections and cropping  Image Adjustments  Measurements      Area and dimensions Histograms ROI and masking Thresholding Quantification of intensity  Macros  Recording and running  Writing your own Additive vs. Subtractive Color CMY additive color RGB additive color K is added to save colored ink  Additive color  Computer monitors, where different intensities of red, green, and blue combine in each monitor pixel to be interpreted by the brain as a specific color. The background of the monitor is black.  Subtractive color  Printer inks absorb different wavelengths from white light.  The background of the image is white (e.g. a sheet of paper)  The primary ink colors are cyan, magenta, yellow, and black, abbreviated CMYK. CMYK example  Subtraction from white  Cyan subtracts red  Magenta subtracts green  Yellow subtracts blue  CMYK is used as this combo produces a wide range of colors from available inks  Why not red, yellow, and blue as primary colors? They produce a narrower range of colors than CMY in modern printing. From: http://www.printingforless.com/rgb-cmyk.html RGB vs. CMYK * * Pantone is a 7-color ink system: CMYK+OGV  CMYK colors cannot be represented in a 1:1 correspondence with RGB values. The space of colors they represent (the “gamut”) do not perfectly overlap.  Some colors in RGB cannot be represented in CMYK.  Which means you can’t print some monitor colors.  Monitor RGB  printer CMYK calibrations are needed!  This usually isn’t noticeable in color photos but can be significant in plots and graphics.

Introduction to ImageJ

Related Documents

Recommended Documents

Description

Related Documents

Free up your schedule!

Take 5 seconds to unlock