| 摘要 |
The proposed method for the subject decoding of an aerospace images consists in determining the allowable values of morphometric indications of the territory, selecting the sheet of the digital map of the territory that covers the specified area of the territory, reading the actual values of the morphometric indications from the sheet, comparing the actual values with the allowable values, and plotting the binary map of the territory. The method is distinctive by determining, after determining the location of each object in the image, the actual values of the morphometric indications of the territory within the area of the location of the object, the parameters that characterize the objects within the territory according to the categories of the objects, the parameters that characterize the similarity of the object to other objects of the same category, and weighting factors for the saidparameters. The data so obtained are used for recognizing objects. The advantage of the method is enhanced accuracy in decoding objects due to the automation if the recognition process.
|
