Research Of Remote Sensing Image Recognition Based On Curve Template

Remote sensing image recognition technology is the basis for the application of remote sensing images, which requires a more rapid and accurate identification technology. At present, remote sensing for different objectives put forward a number of people of different identification methods, have also made good results. But most of the methods have their limitations, for certain types of images may be better to identify the effect, and other types of images has been the effect of a downturn.In this paper, to the low-resolution remote sensing imagery for the study, the design of a universal goal of identifying ways to use images from the template curve detect the existence of the goal, and then from the image quickly and accurately obtain information on the location of the target. The main contribution of this article can be summarized as follows:1 For the low-resolution remote sensing images, a template based on the curve of the image recognition method. As the traditional method of extracting images directly to the edge of the image of higher resolution, does not apply to low-resolution images. Through analysis of the characteristics of the target, using a template extracts curve on the edge of the goal of significantly reducing the image resolution.2 Image binary is an indispensable process to recognition. Traditional binary method are limited, this paper presents a low-resolution images for the "m" word binary methods, and validate the method. In addition, advance a a binary method based on the thinking of classification for precise local binary.3 Design criteria of distinguish goals and reduce interference. As the complexity of the background image, light and shade hopping significant, if not appropriate to select the features, more interference, there is miscarriage of justice may have affected the accuracy of target recognition. In this paper, optimizing the use of the ideas, design criteria objectives, to eliminate as much as possible miscarriage of justice.4 Research of adaptive systems. Different from the high degree of images of objects of different sizes, a high shot on the basis of known information, dynamic adjustment of the curve and the template to determine the criteria of the relevant parameters, so that the proposed method has better adaptability and high positioning accuracy Rate. This paper analyzes the characteristics of the target, to find out the difference between it and the background characteristics of different shapes, and curve fitting using the edge of the goal; the curve in setting the parameters, according to images taken of the distance from the edge of the target image sample analysis to determinethe shape.