All digital images can be described by several characteristics that determine their physical size (the number of bits of memory required to store an image file) and quality. These characteristics are interrelated. For example, the higher the quality of a photo, the usually larger the file size in which it is stored. In order to determine what the quality of a digital image is connected with, it is necessary to get acquainted with such concepts as resolution and graphic formats.
Permission
A digital image is formed from tiny elements called pixels. Pixel is the basic element (building block) of bitmaps. This unit, adopted in computer graphics, similar to the usual meter, kilogram or liter in everyday life. It is the number of pixels in the image that is denoted by the term permission.
The higher the resolution, the more pixels the image contains and, accordingly, the higher the quality of such an image, since an image with a higher resolution is characterized by more detail.
When scanning as well as shooting digital camera or a video camera converts an analog image into digital form (digitization). Currently, sensor devices are mainly used for this purpose.
Sensors are integrated microcircuits in which a set of photosensitive elements is implemented, structurally made in the form of rulers (as in flatbed scanners) or matrices (as in the case of digital cameras). The greater the number of elementary photosensitive elements in the sensor, the greater the resolution it provides.
Sensors with few photosensitive elements do not provide high-resolution images. In such an image, individual elements (pixels) can be seen with the naked eye, which leads to the appearance of steps, i.e. effect pixelation(fig. 2.4).
Conversely, a large number of very small photosensitive elements allow obtaining a digital model of the image close to the original. In the technical documentation for the operation of scanners, the number of dots per inch (dots per inch) is usually used as the units that determine their resolution. That is, when setting the scan mode, you must set scanner resolution in these units, for example, 300 dpi.
NOTE
In the literature, instead of the term dpi (dots per inch), you can find the term ppi (pixels per inch) - pixels per inch. The point is circular and the pixel is square. However, in order to avoid terminological confusion in the future, we will consider the units of measurement of resolution ppi and dpi as synonyms.
Optical (physical) and software (interpolation) resolution
Optical resolution indicates the actual number of photosensitive elements in a square inch (1 inch = 2.54 cm).
Interpolation Resolution is not a physical characteristic of a digital device, but a characteristic of its software. Therefore, the quality of images obtained using interpolated resolution depends on the quality of interpolation algorithms implemented in the program.
For example, a scanner passport may contain an optical resolution of 1200 dpi, and a software resolution of 24000 dpi ..
NOTE
Many professional photographers have a negative attitude towards increasing the resolution of photographic images not by hardware, but by software, since when the resolution is reduced, the data is discarded, and when the resolution is increased, the program "invents" it. In other words, interpolation artificially adds elements of a digital image, but does not increase the amount of image detail.
Monitor Resolution
The resolution of a monitor is related to the maximum number of dots it can generate and their size, and is measured by the number of dots in one horizontal line and the number of horizontal lines on the screen. With today's typical dot size ("grain") of 0.2 mm, the standard resolution for 17-inch monitors is 1024x768.
Printer Resolution
The resolution of a laser printer is determined by the number of dots that the printer can print on one inch (dpi - dots per inch). So, if a laser printer has a resolution of 300 dots per inch, then it can print 300 dots in one inch.
You can see the resolution of the printer you have installed by running the command Start Control Panel Printers and Faxes (Fig. 2.5).
Rice. 2.5.
Digital camera resolution
In a digital camera, light passing through the lens enters the light-sensitive matrix (which takes the place of the film) - a set of CCD (CCD) or CMOS (CMOS) sensors, which digitize the image. In the process of digitizing an image from a digital camera, the information contained in it is converted into a set of numbers organized in a matrix called bit matrix(bit-map). In this case, each photocell of the sensor corresponds to a certain numerical element in the bit matrix.
The light sensor (sensor) is the main (and most expensive) component of a digital camera. The quality of the image taken by the camera depends mainly on the resolution of the sensors and the quality of the optics of the camera.
In digital cameras, the basic unit of measurement for resolution is the pixel, and its value is determined by the size of an individual cell of the CCD matrix.
For images entered into a computer using a digital camera, the resolution can be specified either as a specific number of megapixels (a megapixel sensor contains 1 million photosensitive cells) or as a raster image with a specified number of pixels horizontally and vertically. For example, a digital camera with a 2.1 megapixel touchscreen creates an image file of 1792 * 1200 pixels (saved in JPEG format).
Graphics formats
After the frame is taken in a digital camera, the resulting image must be written into memory. For this, graphic formats are most often used. Jpeg or Tiff... Moreover, for a photographer, the recording format is not so much important as the capabilities of the compression modes used in them (preferably with a minimum loss of quality), as well as the amount of memory in the camera. Let's talk about this in more detail.
Each of the formats existing today has passed natural selection, has proven its viability and practical value. All of them have characteristic features and capabilities that make them indispensable in specific areas of application: Web-design, printing, photo retouching and others.
All the many formats used for recording images can be roughly divided into two categories:
- storing images in raster form (BMP, TIFF, JPEG, PNG, GIF, etc.);
- storing the image in vector form (WMF, CDR, AI, FH9, etc.);
Which format should you give preference to? Professionals know that it is better to save work results in a format that is "native" for the program being used. For example, in Photoshop - PSD, CorelDRAW - CDR, Flash - FLA. This will allow in maximum degree to realize the possibilities of the program and to insure against unpleasant surprises. However, in this lecture we will focus mainly on raster formats, since you have to work with photography in raster formats. graphic editors.
Raster formats
A raster image (raster) resembles a grid (table) of pixels, which in the simplest black and white version consists of two types of cells: white or black, and which can be encoded, respectively, zero or one. Unlike black-and-white, in a color RGB image, for example, 24-bit deep, each pixel is encoded with a 24-bit number, therefore, a number of 24 ones and ones is stored in each cell of the bit matrix.
Now let's move on to looking at the most common bitmap formats.
BMP
The BMP format (from the word bitmap) is a native Windows format. It is supported by all graphics editors running this operating system. It is used to store bitmap images intended for use in Windows, for example, as the background of your desktop. With this format, you can set the color depth from 1 to 24 bits. Provides the ability to apply data compression by the algorithm
Angular resolution- the minimum angle between objects that the optical system can distinguish.
The ability of an optical system to distinguish points of the imaged surface for example:
Angular resolution: 1 ′ (one arc minute, about 0.02 °) corresponds to a 29 cm area, distinguishable from a distance of 1 km or one printed point of the text at a distance of 1 m.
Linear Resolution
General information
The resolution of optical devices is fundamentally limited by diffraction on the lens: the visible points are nothing more than diffraction spots. Two adjacent points are resolved if the minimum intensity between them is small enough to be seen. To remove the dependence on the subjectivity of perception, an empirical criterion Rayleigh permission which defines the minimum angular distance between points
sin θ = 1.22 λ D (\ displaystyle \ sin \ theta = 1.22 (\ frac (\ lambda) (D)))where θ - angular resolution (minimum angular distance), λ - wavelength, D- the diameter of the entrance pupil of the optical system (often it coincides with the diameter of the objective). Given the extreme smallness of the angle θ , in the optical literature, instead of the sine of the angle, the angle itself is usually written.
The coefficient was selected so that the intensity at the minimum between the spots was equal to approximately 0.75-0.8 of the intensity at their maxima - it is believed that this is sufficient for discrimination with the naked eye.
The dependence of the resolution when photographing on the properties of the optical system
When photographing for the purpose of obtaining a print or image on a monitor, the total resolution is determined by the resolution of each stage of object reproduction.
Methods for determining the resolution in photography
Resolution is determined by photographing a special test object (targets). To determine the resolution of each of the elements involved in the technical process of image acquisition, measurements are carried out under conditions when the errors from the remaining stages are negligible.
Resolving power of the lens
Resolution of the primary material medium
Photographic emulsion
It is important that the modern foreign interpretation lines worlds counts a couple black and white stripe- per 2 lines, in contrast to domestic theory and practice, where each line it is always considered to be separated by the intervals of the contrasting background with a thickness equal to the thickness of the line.
Some companies - manufacturers of digital cameras for advertising purposes are trying to rotate the matrix at an angle of 45 °, achieving a certain formal increase in resolution when photographing the simplest horizontal-vertical world. But if you use the professional world, or at least turn the simple world at the same angle, it becomes obvious that the increase in resolution is fictitious.
Getting the final image
The resolution of modern printers is measured in dots per millimeter (dpmm) or per inch (dpi).
Inkjet printers
The print quality of inkjet printers is characterized by:
- Printer resolution (DPI unit)
- The color resolution of the system printer-ink-color ICC profiles (print color fields). The color fields of the print are largely limited by the properties of the ink used. If necessary, the printer can be converted to almost any ink that matches the type of printheads used in the printer, and it may be necessary to reconfigure the color profiles.
- Resolution of the printed image. Usually it differs very much from the resolution of the printer, since printers use a limited number of colors, maximum 4 ... 8 and mosaic color mixing is used to obtain halftones, that is, one image element (analogue of a pixel) consists of many elements printed by the printer (dots are ink drops)
- The quality of the printing process itself (material movement accuracy, carriage positioning accuracy, etc.)
To measure the resolution of inkjet printers, in everyday life, a single unit of measurement is adopted - DPI, which corresponds to the number of dots-physical drops of ink per inch of the printed image. In reality, the actual resolution of an inkjet printer (apparent print quality) depends on many more factors:
- In most cases, the printer control program can work in modes that provide very slow movement of the print head and, as a result, with a fixed frequency of ink spraying by the nozzles of the print head, a very high “mathematical” resolution of the printed image is obtained (sometimes up to 1440 × 1440 DPI and higher). However, it should be remembered that the real image does not consist of "mathematical" points (infinitely small diameter), but of real paint drops. At an unreasonably high resolution, more than 360 ... 600 (approximately) the amount of ink applied to the material becomes excessive (even if the printer is equipped with heads that create a very small drop). As a result, to obtain an image of a given chromaticity, the fill has to be limited (that is, to return the number of drops of paint to reasonable limits). For this, both pre-made settings that are embedded in the ICC color profiles and a forced decrease in the percentage of filling are used.
- When printing a real image, the nozzles are gradually blocked by internal factors (air bubbles entering with ink entering the print head nozzles) and external factors (adhesion of dust and ink droplets on the surface of the print head). As a result of gradual blocking of nozzles, unprinted stripes appear on the image, the printer begins to "strip". Nozzle blocking speed depends on the type of print head and carriage design. The problem of clogged nozzles is solved by cleaning the print head.
- The nozzles do not spray down perfectly, but have a small angular spread depending on the type of printhead. Droplet drift due to scatter can be compensated for by reducing the distance between the printhead and the printed material, but remember that a head that is too lowered can catch the material. Sometimes this leads to rejects; with particularly hard hooks, the print head can be damaged.
- The nozzles in the print head are arranged in vertical rows. One row - one color. The carriage prints both when moving from left to right and from right to left. When moving in one direction, the head is the last to put one color, and when moving to the other side, it is the last to put another color. The paint of different layers, falling on the material, is only partially mixed, there is a color fluctuation, which looks different in different colors. Somewhere it is almost invisible, somewhere it is strongly striking. Many printers have the ability to print only when the head moves in one direction (to Left or to Right), the reverse is idle (this completely eliminates the "mattress" effect, but greatly reduces the print speed). Some printers have a double set of heads, while the heads are arranged in a mirror (example: Yellow-Pink-Cyan-Black-Black-Cyan-Pink-Yellow), such an arrangement of the heads excludes the effect under consideration, but requires more complex adjustment - convergence of heads of the same color between themselves.
Laser and LED printers
Monitors
It is measured in points per unit of image length on the surface of the monitor (in dpmm or dpi).
Microscopes
the optical medium in which the lens is located. λ - the wavelength of light illuminating or emitted by an object (for fluorescence microscopy). Meaning n sin α also referred to as numerical aperture.Due to overlapping value constraints α , λ , and η , the resolution limit of the light microscope, when illuminated with white light, is approximately 200 ... 300 nm. Insofar as: α the best lens is approximately 70 ° (sin α = 0.94 ... 0.95), considering also that the shortest wavelength of visible light is blue ( λ = 450 nm; purple λ = 400 ... 433), and typically high resolutions are provided by the lenses of oil-immersion objectives ( η = 1.52 …1.56 ; by I. Newton 1,56 is the refractive index (index) for violet), we have:
R = 0.61 × 450 nm 1.56 × 0.94 = 187 nm (\ displaystyle R = (\ frac (0.61 \ times 450 \, (\ mbox (nm))) (1.56 \ times 0.94)) = 187 \, (\ mbox ( nm)))For other types of microscopes, the resolution is determined by different parameters. So, for a scanning electron microscope, the resolution is determined by the diameter of the electron beam and / or the diameter of the region of interaction of electrons with the sample substance.
A scanner is a device that allows you to graphically enter text, drawings, slides, photographs, graphs, articles, manuscripts, etc. into your computer. All scers can be divided into several classes: hand-held (extended), desktop or tablet, scanners for transparent materials. The main differences between the devices are cost, image quality and method of use.
Scanners are SAD systems (Source Attenuator Detector). When the light in the scanner bounces off or passes through the document, the amplitude of the light signal will decrease slightly, which will register the sensors on the scanner, which measure the difference between the light values. There are different types of sensors. Most scers use Charged-coupled devices (charge coupled devices), or charge coupled devices (CCDs) that convert light into piels. Each scanner has a linear array of several thousand CCDs arranged in a row along the scanning engine. Some scanners use Complementary Metal-Oxide Semiconductor (CMOS) sensors, which first appeared in digital cameras. CMOS devices differ from CCD sensors in that they exist as a separate unit. CCD and CMOS devices during scanning compare the amount of electric charge before and after it is reflected from the scanned original. The difference is converted to hue and determines the color of the pixels.
Scanning speed is one of the characteristics of the scanner.
The scan time starts at the push of a button. Scan and will end with the moment when the image is available for editing in Adobe Photoshop. If scanning is performed with the autocalibration mode turned on, which is performed before each scan, then the scanning time increases by 6-8 s.
Research shows that scan time with resolutions of 1200 and 2400 dpi, it turned out to be the same, which suggests that the vertical resolution, which many manufacturers recently declare for advertising purposes twice as large as horizontal, is most likely just an interpolation resolution, and the figure 2400 only indicates improved scanner mechanics.
Modern scanners have a fairly large memory buffer: when scanning 50 MB pictures while the ruler is parked (movement of the scanners to the initial position), the scanner continues to calculate and transmit the image.
Dynamic range- one of the most important parameters of the scanner. The dynamic range is calculated by the formula: D = Dmax - Dmin, where D is the difference between the maximum and minimum optical densities distinguished by the scanner. Usually, the minimum optical density Dmin, perceived by the scanner, is 0.07-0.08 D.
Optical density equal to the negative decimal logarithm of the reflection (transmission) coefficient. If the optical density is 1, 2, 3, etc., then one tenth, one hundredth or one thousandth of the incident light is reflected (or transmitted), respectively. On transparent graphic materials (slides) and photographic images, the optical density can reach 4.0.
Scanner optical resolution
The main characteristic of the scanner is optical resolution... It is measured in ppi - pixels per inch, often, however, dpi - dots per inch. The term "point" means an element that does not have a specific shape and is sweet for measuring the resolution of printing devices. Scanners and bitmap graphic files operate with pixels that are always square.
Optical resolution indicates how many pixels the scanner can read in a square inch. It is written as follows: 300´300, 300´600, 600´1200, etc. The first number indicates the number of sensors reading information, it is worth paying attention to it. Often manufacturers and sellers like to specify something like 4000, 4500 dpi in the resolution value. This interpolated solution is not a property of the scanner, but of the program that supports it. The quantity of images obtained in this way depends not only on the scanner, but also on the quality of the interpolation functions implemented in the program.
Of course, scanning the maximum range of optical densities is not at all necessary, and sometimes not desirable - with normal, not test scanning.
Another unit for measuring optical resolution is spi (samples per inch) - the number of samples taken by the scanner in one inch. In this case, the resolution indicates how many times the scanner scans the image when scanning. If there are 600 tiny sensors in a line array of a flatbed scanner every inch in a row, then the optical resolution of the scanner is 600 spi.
Optical resolution in dpi is usually indicated by scanner manufacturers, although it is more logical to specify in spi.
Hello dear blog readers about. Today we will talk about such an important scanning parameter as permission... Resolution determines the amount of detail recorded. It is measured in dots per inch (dpi). The higher the dpi value, the higher the resolution.
The quality of the image improves with the increase in resolution, but only until a certain point, after which a further increase in resolution leads only to the fact that the file becomes too large to be manipulated. Plus, higher resolution images take longer to print. In most cases, 300 dpi is more than sufficient for scans.
Talking about scanner resolution Don't forget about the difference between optical resolution and interpolation. Optical resolution is native to the scanner and depends on the optics used in the design of the device. The interpolated resolution is the resolution increased with special software. And although interpolation can be useful in some cases (for example, when scanning graphics or when you need to enlarge a small image), the quality and clarity of the image obtained in this way is lower than when using only optical resolution.
How do I choose the best resolution settings?
Scanning at high resolution takes more time, memory, and disk space. When making resolution settings, take into account the type of image and printing method that you intend to use in the future, or the output device.
The easiest way to determine the required resolution is to find out the number of lines per inch (lpi value) of the image output device and, to be more accurate, multiply this number by two.
Example: To "fit" your scanned image to fit a standard magazine press with a lpi of 133, simply multiply 133 by 2. The result is an optimal resolution of 266 dpi. However, if you intend to enlarge the image after scanning, remember that this will decrease the resolution, so be careful with scaling.
The lpi varies depending on the print quality. A newspaper needs about 85 lpi, a magazine needs 133-150 lpi, and a color book may need 200 to 300 lpi.
If you are displaying images on a monitor (for example, for publishing on the Internet), there is no need for a resolution of more than 72 dpi, since monitors are not capable of displaying more than 72 dpi. Larger resolution images will not get better or clearer; it will only increase the size of the file, making it more difficult to process.
Remember that the higher the resolution, the larger the file size. For example, a color photo of 8.5 by 11 inches with a resolution of 72 dpi will "weigh" about 1.6 megabytes. Increasing the resolution to 150 dpi will increase the file to 6.3 megabytes (about four times)! And at 300 dpi the same file will already "weigh" 26.2 megabytes.
Thus, you should always try to choose the lowest possible resolution in order to maintain image quality and at the same time, get a file that is not too large for easy use.
When do you need high resolution?
High resolution is important when you run your image through a high-tech color management system that preserves all the data captured during the scan when printed. In this case, high resolution will make the final image clearer and sharper.
When to use an interpolated image?
The interpolation function is useful for scanning graphics and pencil drawings, and for enlarging small sized images. Also included in this category are any black and white or monochrome graphics, ink or pencil sketches, sketches, or mechanical blueprints.
For graphics: set the resolution equal to the resolution of the printing device. For example, if you intend to print an image on a 1200 dpi device, set the scanner to 1200 dpi for optimal results. This will make the lines smoother and eliminate bumps and fuzziness.
To enlarge small originals: Suppose you are scanning a 1-inch or 2-inch photograph at 300 dpi, and the maximum optical resolution of the scanner is also 300 dpi. To enlarge the image twice the size of the original without losing detail, interpolate the image up to 600 dpi. Thus, the image will remain sharp and clear, and its size will be doubled.
Page 1
Optical resolution is defined as the number of light-sensitive elements in the scanning head divided by the width of the working area. As a rule, high resolution is necessary only for comfortable visual perception.
Optical resolution is a parameter that determines the maximum resolution of the scanner without interpolation. Resolution is usually measured by the number of separately fixed or observed image points of an object per unit of fixed length or area.
For example, if the optical resolution of the scanner is 800 dpi, then the manufacturers write in the documentation in large letters that the scanner resolution is 1200 dpi or even 1600 dpi, while adding in small letters that this is the software resolution.
Due to the extremely short wavelength, an optical resolution is achieved that is much greater than when using visible light. The resolution limit for an electron microscope is a few tenths of a nanometer; in visible light, only about a thousand times larger details can be discerned.
These conditions mean optical resolution in an electron microscopic image, sufficient to distinguish polygonal outlines, for example, in particles of gold hydrosols.
In the development of this camera, large investments have been made to achieve both the highest possible optical resolution and extremely short exposures.
Phil's beautiful drawings were fine-tuned by Robin Ruskin and then printed on a QMS Lasergrafix at four times the original optical resolution. The quality of these drawings was not as high as a professional artist could achieve with pen and ink, but the possibility of additional personal involvement seemed more important to me.
It is necessary to distinguish between the concepts of optical resolution of the scanner and software resolution. Optical resolution indicates the limiting capabilities of the scanner's photosensitive element. However, with the help of software tweaks, that is, using special algorithms for working with images, you can increase the resolution of the scanner, usually to the next value in the above series.
Generally speaking, at low concentrations it can be compared with the color of a dichromate solution, but this is not a general rule. The problem of optical resolution of such liquids can be solved only with a spectrocolorimeter. Such devices have not yet entered the oil practice and to some extent they are being replaced by the Lovibond tintometer.
Resolution is determined by the number of optical elements per unit length. Software interpolation of optical resolution does not give any real improvement in the quality of digitization. The dynamic range of CCD devices is lower than that of PMTs because silicon elements have a poorer signal-to-noise ratio.
These values determine the theoretical limit of the optical resolution of photolithography.
In most preparations, the local transmission varies from point to point across the microscope field of view. The influence of this inhomogeneity leads to the appearance of a specific error - averaging error. This error always occurs when examining regions of the specimen with different transmissions, the size of which is greater than the optical resolution limit.
A thin disc of germanium, with a concave surface on one side, is anchored by wire contact on this surface. The other side of the germanium disc is exposed to light, which can be focused on it using a lens. When an appropriate polarizing voltage is applied, the output current of the phototransistor depends on the illumination. Since the active surface of the photocell is very small, good optical resolution is obtained. The spectral characteristic covers both the visible light and infrared range up to wavelengths of about 2 microns with a maximum of about 15 microns. The phototransistor has a fairly flat frequency response up to 200 kHz. It is possible to obtain an output current of 0 07 ma / mlm even with a load of 2 kΩ.
A photographic lens, placed between the photographic plate and the sample, focuses the image of the object's surface onto the plane of the photographic plate. Moreover, their planes should be parallel. A significant advantage of focused image holograms is the ability to obtain an enlarged image of an object, and, consequently, a higher optical resolution of interference fringes.
Some types of cementite, for example, tertiary cementite or cementite distributed in the structure of steels after hardening, are detected with this etchant better than with etchants, after which the iron carbide looks dark against the background of the surrounding light matrix. Klemm used it to detect cementite and the γ-phase in hardened structures. Etching does not require removing the deformed layer of the ferrite matrix. The image of the structure is obtained with a better quality if the sulfide deposit on the entire surface of the ferrite is equally oriented. This method allows you to monitor the development of coagulation of cementite released during the tempering process. Naturally, for the study of a small number of the smallest particles of cementite, optical resolution is of great importance.
