基于数据块特征的地面目标识别方法研究

Image Recognition Way Based on Data Block Character

光线变化常常改变图像数据块结构,导致图像匹配识别错误。文中分析了光线强弱对地面目标灰度变化影响,提出基于目标数据块特征的识别方法,将数据块目标分为反色或不反色两类,分别用Nprod匹配和特征识别方法识别两类目标,避免了光线变化对图像识别影响。实验表明,该方法能够克服光线变化影响,较好地识别地面不同类型目标。

Light ray changes can always alter the image data block structure ,which leads to matching and recognition errors in the image. This paper analyzes how the intensity of light impacts on the gray of ground targets, and proposes a kind of method to recognize the target based on data block as well as dividing the target into inverse color or direct color part. We used the Nprod matching and character distin- guishing method to extract the target to avoid the image gray disturbing by the ray. It is proved that the method we put up with can overcome the impact of changes in light and recognize different kinds of targets aground well.