How are pixels calculated?

"Pixel" consists of letters of the words "picture" and "element", which is the unit used to calculate digital images. Just like photos, digital images have continuous shadows. If you enlarge the image several times, you will find that these continuous tones are actually composed of many small squares with similar colors, which are the smallest unit "pixels" that make up the image. This smallest graphic unit can usually display a coloring point on the screen. The higher the pixel, the richer the sample and the more realistic the color is.

Pixel is usually considered as the smallest complete sample of an image. This definition is very relevant to the context. For example, we can say that pixels in a visible image (such as a printed page) are pixels represented by electronic signals, pixels represented by numbers, pixels on a display or pixels in a digital camera (photosensitive element). Many other examples can be added to this list. Depending on the context, there will be some more accurate synonyms, such as pixels, sampling points, bytes, bits, dots, spots, supersets, triples, edge sets, windows and so on. We can also discuss pixels abstractly, especially when using pixels as a measure of resolution, such as 2400 pixels per inch (ppi) or 640 pixels per line. Dots are sometimes used to represent pixels, especially computer marketers, so ppi is sometimes written as DPI (dots per inch).

The more pixels used to represent an image, the closer the result is to the original image. Although there is a more specific definition of resolution, the number of pixels in an image is sometimes called image resolution. Pixels can be represented by a number, such as a "3 megapixel" digital camera with a rated pixel of 3 million, or by a pair of numbers, such as a "640 × 480 display" with 640 horizontal pixels and 480 vertical pixels (just like a VGA display), so its total number is 640× 480 = 307,200 pixels.

The color sampling points of digital images (such as JPG files commonly used in web pages) are also called pixels. Depending on the computer monitor, these pixels may not have a one-to-one correspondence with the screen pixels. In this obviously different area, the points in the image file are closer to the texture elements.

In computer programming, an image composed of pixels is called a bitmap or raster image. Grating once originated from analog TV technology. Bitmap images can be used to encode digital images and some types of computer-generated art.

Original and logical pixels

Because the resolution of most computer monitors can be adjusted by the computer's operating system, the pixel resolution of the monitor may not be an absolute measure.

Modern liquid crystal displays have original resolution by design, which represents the perfect match between pixels and triplets. Cathode-ray tubes also use red, green and blue fluorescent triplets, but they do not coincide with image pixels, so they cannot be compared with pixels.

For this display, the original resolution can produce the finest image. But because the user can adjust the resolution, the display must be able to display other resolutions. Non-original resolution must be achieved by fitting resampling on the LCD screen, and interpolation algorithm should be used. This often makes the screen look broken or blurred. For example, the display with the original resolution of1280x1024 looks the best. A pixel represented by several physical triplets can also display 800×600, but it may not be able to display 1600× 65434 completely.

Pixels can be rectangular or square. There is a number called aspect ratio, which means there are many pixels. For example, the aspect ratio of 1.25: 1 means that the width of each pixel is 1.25 times its height. Pixels on computer monitors are usually square, but pixels used for digital images have rectangular aspect ratios, such as those used in PAL and NTSC, CCIR 60 1 digital image standards and corresponding widescreen formats.

Each pixel of a monochrome image has its own brightness. 0 usually means black, and the maximum value usually means white. For example, in an 8-bit image, the maximum unsigned number is 255, so this is a white value.

In a color image, each pixel can be represented by its hue, saturation and brightness, but it is usually represented by the intensity of red, green and blue (see red, green and blue).

Bit per pixel

The number of different colors that a pixel can express depends on the number of bits per pixel (BPP). This maximum number can be obtained by taking the square of color depth. For example, common values are:

8 bpp[28 = 256； (256 colors)];

16 bpp[2 16 = 65536； (65536 colors, called high color)];

24 bpp[224 = 167772 16； (16,777,216 color, called true color)];

48 bpp[248 = 28 14749767 10656； (28 1, 474, 976, 7 10, 656 colors, which are used by many professional scanners).

Graphics of 256 colors or less are usually stored in a block or plane format in a video memory, where each pixel in the video memory is an index value of a color array called a palette. Therefore, these patterns are sometimes called index patterns. Although there are only 256 colors at a time, these 256 colors are selected from a much larger palette, usually 16 megacolors. You can get the animation effect by changing the color values in the palette. Logos of Windows 95 and 98 are probably the most famous examples of this kind of animation.

For depths exceeding 8 bits, these numbers are the sum of three components (red, green and blue). The depth of 16 is usually divided into 5 bits of red, 5 bits of blue and 6 bits of green (eyes are more sensitive to green). The depth of 24 bits is usually 8 bits per component. In some systems, 32-bit depth is also optional: this means that a 24-bit pixel has 8 extra numbers to describe transparency. In the old system, 4bpp( 16 color) is also very common.

When an image file is displayed on the screen, the number of bits per pixel may be different for raster text and display. Some raster image file formats have greater color depth than others. For example, the maximum depth of GIF format is 8 bits, while TIFF files can handle 48-bit pixels. No display can display 48-bit colors, so this depth is usually used for special professional applications, such as film scanners and printers. This file is drawn at a depth of 24 bits on the screen.

Subpixel

For different reasons, many monitors and image acquisition systems cannot display or perceive different color channels at the same point. This problem is usually solved by multiple sub-pixels, and each sub-pixel handles one color channel. For example, LCD displays usually divide each pixel into three sub-pixels in the horizontal direction. Most LED displays decompose each pixel into 4 sub-pixels; One is red, one is green and two are blue. Most digital camera sensors also use sub-pixels and are realized by color filters. (CRT monitors also use red, green and blue fluorescent dots, but they are not aligned with image pixels, so they cannot be called sub-pixels).

For systems with sub-pixels, there are two different processing methods: sub-pixels can be ignored and regarded as the smallest accessible image element, or sub-pixels can be included in rendering calculation, which requires more analysis and processing time, but in some cases, better images can be provided.

The latter method is used to improve the appearance resolution of color display. This technique, called sub-pixel rendering, uses pixel geometry to manipulate sub-pixels individually, which is most effective for flat panel displays with original resolution (because the pixel geometry of such displays is usually fixed and known). This is a form of anti-aliasing, mainly used to improve the display of text. Microsoft's ClearType, which can be used on Windows XP, is an example of this technology.

Megapixel

Megapixels are one million pixels, which are usually used to represent the resolution of digital cameras. For example, a camera can use a resolution of 2048× 1536 pixels, which is usually called "31million pixels" (2048×1536 = 3145728).

Digital continuously uses photosensitive electronic devices, or coupled charge devices (CCD) or CMOS sensors to record the brightness level of each pixel. In most digital cameras, CCD adopts a certain arrangement of color filters, and there are three areas of red, green and blue in Bayer filter module, so that photosensitive pixels can record the brightness of a single primary color. The camera interpolates the color information of adjacent pixels, a process called demosaicing, and then creates the final image. In this way, the final color resolution of an x-megapixel image in a digital camera may be only a quarter of that of the same image in a scanner. In this way, the images of blue or red objects tend to be more blurred than those of gray objects. Green objects seem to be less blurred because green is assigned more pixels (because eyes are sensitive to green). See [1] for a detailed discussion.

As a new development, Foveon X3 CCD uses a three-layer image sensor to detect the red, green and blue intensities of each pixel. This structure eliminates the need for spelling, thus eliminating related image aliasing, such as color blur at high contrast edges.

Similarity concept

Several other concepts are derived from the concept of pixels, such as voxels, texture elements and surface elements, which are used in other computer graphics and image processing applications.

Pixel of digital camera

Pixel is the most important index to measure a digital camera. Pixel refers to the resolution of a digital camera. Depending on the number of photosensitive elements on the photoelectric sensor in the camera, one photosensitive element corresponds to one pixel. Therefore, the larger the pixel, the more photosensitive elements, and the greater the corresponding cost.

The image quality of a digital camera is determined by pixels. The larger the pixel, the greater the resolution of the photo, and the larger the print size without reducing the print quality. Early digital cameras were all below 6.5438+0 million pixels. Since the second half of 1999, 2-megapixel products have gradually become the mainstream of the market. At present, the development trend of digital cameras, pixels are like the main frequency of a PC's CPU, and there is an increasing momentum.

In fact, from the perspective of market classification, for popular products, considering the factor of cost performance, the bigger the pixel, the better. After all, a 2-megapixel product can meet most applications of ordinary consumers. For this reason, while most manufacturers pursue high-end digital cameras with high pixels, the largest output at present is still the popular mega-pixel products. Professional digital cameras have products with more than 1 100 million pixels. With the progress of CCD (imaging chip) manufacturing technology and the further reduction of cost, 3 million pixel products will soon become the mainstream of the consumer market.

In addition, it is worth consumers' attention that the current digital camera products are nominally divided into CCD pixels and software-optimized pixels, and the latter is much higher than the former. For example, a brand of Volkswagen digital camera has 2.3 million CCD pixels, while the pixels after software optimization can reach 3.3 million.

Pixel painting

Pixels are actually made up of many points.

The "pixel painting" we are talking about here is not a dot matrix image corresponding to a vector diagram, but an icon image. This style of image emphasizes clear outline and bright color, and the shape of pixel map is often cartoon, so it is loved by many friends.

The method of making pixel map hardly uses aliasing to draw smooth lines, so it is often used. Gif format, and pictures often appear in dynamic form. However, due to its special production process, it is difficult to guarantee the style if the size of the picture is changed at will.

Pixel painting is widely used, from the picture of FC home red and white machine played as a child to GBA handheld machine today; From black and white mobile phone pictures to today's full-color handheld computers; Even the computer we face every day is full of pixel icons of various software. Nowadays, pixel painting has become an art, which deeply shocks you and me.

Effective pixel value

First of all, it should be clear that the actual pixel values of digital photos are different from those of sensors. Take a general sensor as an example, each pixel has a photodiode, which represents a pixel in the photo. For example, a 5-megapixel digital camera, whose sensor can output an image with a resolution of 2560 x 1920-in fact, this value is only equivalent to 4.9 million effective pixels. Other pixels around the effective pixel are responsible for other tasks, such as deciding "what is black". Many times, not all pixels on the sensor can be used. Sony F505V is one of the classic cases. The sensor of Sony F505V has 3.34 million pixels, but it can intelligently output 1, 856 x 1, 3.92 or 2.6 million pixels at most. The reason is that Sony put a new sensor bigger than the old one in the old digital camera at that time, which caused the sensor to be too big, and the original lens completely covered every pixel in the sensor.

Therefore, the digital camera uses the principle that the pixel value of the sensor is greater than the effective pixel value to output digital pictures. In today's market, which is constantly pursuing high pixels, digital camera manufacturers often target sensor pixels with higher values in advertisements, rather than effective pixels that reflect the actual imaging clarity.

Sensor pixel interpolation

Generally speaking, each pixel in different positions in the sensor constitutes each pixel in the picture. For example, a photo with 5 million pixels is obtained by measuring and processing the light entering the shutter with 5 million pixels in the sensor (other pixels except the effective pixels are only responsible for calculation). But sometimes we can see such a digital camera: it only has 3 million pixels, but it can output 6 million pixels of photos! In fact, there is nothing false here, but the camera calculates and interpolates on the basis of the sensor's 3 million pixel measurement to increase the number of pixels in the photo.

When photographers take photos in JPEG format, the quality of this "built-in magnification" will be better than that of our computer, because the "built-in magnification" is completed before the pictures are compressed into JPEG format. Photographers who have experience in digital photo processing know that enlarging JPEG images in a computer will make the picture quick, delicate and smooth. Although the image quality of digital camera interpolation is better than the normal output of sensor pixels, the file size of the interpolated image is much larger than the normal output image (for example, 3 million sensor pixels are interpolated into 6 million pixels, and the image finally input into the memory card is 6 million pixels). Therefore, the high pixels obtained by interpolation do not seem to have much merit. In fact, using interpolation is just like using digital zoom-it can't create details that the original pixels can't record.

CCD total pixel

The total pixel of CCD is also a very important index. Because different manufacturers adopt different technologies, the nominal CCD pixels of their manufacturers do not directly correspond to the actual pixels of the camera, so it is more important to look at the actual total pixels of the camera when buying a digital camera. Generally speaking, if the total pixel level reaches about 3 million, it can meet the general application. Generally, products with 2 million pixels and 6.5438+0 million pixels can also meet the needs of low-end use. Of course, digital cameras with higher pixels can get higher quality photos. Now some companies have begun to introduce 600 million pixel ordinary digital cameras.