Pixel is the smallest unit of Raster Data, and it is also an important symbol to reflect the image characteristics. In remote sensing Data Collection, such as scanning imaging, the pixel is the minimum unit for the sensor to scan and sample the ground scene. Pixel is a data element with both spatial and spectral characteristics. The spatial characteristic is that the pixel represents the ground area; the physical meaning is that the spectral variable represents the intensity of the spectral response of the pixel in a specific band. Objects in the same pixel have only one common pixel value.

Pixel size

Pixel size determines the resolution and information content of Raster Data. If the pixel is small, the spatial resolution of the image is high, and the Details of the ground objects can be obtained; if the pixel is large, the resolution of the image is low, and the amount of information contained in the pixel is small, so that the execution efficiency of computer storage and analysis can be improved. For example, the pixel size of Landsat MSS image is 56 × 79 square meters, and the number of single-band pixels is 7581600, while the pixel size of TM image is 30 × 30 square meters, and the number of single-band pixels is 38023666, which is equivalent to 5 times of MSS data.

Although the smaller the pixel is in the same element range, the more detailed the object features and information are, the smaller the pixel is, the better it is, which depends on the user's needs and computer performance. If the pixel is smaller, the Raster Dataset will be larger when representing the feature information in the same range, so it will require more storage space, higher computer performance requirements, and longer Data Processing and analysis time.

Pixel value

In Raster Dataset, each pixel (pixel) has an attribute value, and the pixel has a certain spatial resolution, that is, it corresponds to a certain area of the earth's surface, so the pixel value represents the dominant element or phenomenon of the area covered by the pixel. For example, the spectral values in Satellite Image and aerial photos reflect the reflectance of light in a certain band; the elevation values of the DEM grid represent the surface elevation above the mean sea level, and the pixel values of the slope map, aspect map and watershed map generated by the DEM grid represent their slope, aspect and watershed attributes, respectively; The category values in the Land Use classification map, such as cultivated land, forest land, grassland, etc., can also represent the quantitative values of precipitation, temperature, pollutant concentration, distance, etc. In addition, the pixel value can be an integer or a floating point number.

No value

When some pixel values are missing or meaningless data, no value can be used as the value of the pixel. A null value is generally identified by a less common, more specific numeric value. In SuperMap, the value of a null value is usually specified as -9999. Note: Valueless data is not equal to 0. 0 is a valid value.

In the grid analysis function, the processing of no value is generally different from that of other pixel values. There are three ways: Ignore No Value, in which no value is not involved in the operation; Calculate Result in the no value area is still no value; and estimate the value of no value data. Valueless data is typically handled differently in different Raster Analysis operations. For example, when making Grid Neighborhood Statistics, there is no value data around the grid cell to be calculated. In this case, you can select two processing methods. You can ignore the no value data and use other valid values to calculate, or you can not ignore it. Output Result is no value. When the former method is adopted, the Calculate Result is not necessarily correct, because the ignored valueless data is likely to be the minimum value or the maximum value in the neighborhood.

Resolution

In raster and Image Data, there are four types of Resolution: spatial resolution, temporal resolution, spectral resolution and radiometric resolution.

• Spatial resolution: Spatial resolution, also known as pixel size, is the size of the area covered on the ground represented by a single pixel, in meters or kilometers. For example, one pixel of American QuickBird commercial Satellite Image is equivalent to the ground area of 0.61 m × 0.61 m, and its spatial resolution is 0.61 m; one pixel of Landsat/TM multi-band image covers the ground area of 28.5 m × 28.5 m, and its spatial resolution is 28.5 m. When

  representing the same size area of the earth's surface, the image with high spatial resolution needs more pixels than image with low spatial resolution, that is, the grid with smaller pixel size needs more rows and columns to represent, so that more information and details of the earth's surface can be displayed. Therefore, the higher the spatial resolution is, the more the details of the surface are stored, the larger the storage space is, and the longer the Data Processing time is; on the contrary, the lower the spatial resolution is, the coarser the surface information is, but the storage space is smaller, and the processing speed is faster. Therefore, when choosing the pixel size, that is, spatial resolution, it is necessary to take into account the requirements of practical applications for the level of detail of information, as well as the requirements for storage and Data Processing time and speed.

• Time resolution: It refers to the minimum time interval between two adjacent remote sensing observations in the same area. When the time interval is large, the time resolution is low; conversely, when the time interval is small, the time resolution is high.
• Spectral resolution: refers to the imaging band range, the finer the division, the more bands, the higher the spectral resolution. Generally speaking, the more bands the sensor has and the narrower the band width is, the easier it is to distinguish and identify the information of ground objects.
• Radiometric resolution: It refers to the minimum variation of electromagnetic radiation intensity reflected or radiated by the target that can be resolved by the sensor. Describes the ability of the sensor to resolve the object being viewed in the same portion of the electromagnetic spectrum.