论文部分内容阅读
农作物与成像系统之间的相对运动,造成了图像的运动模糊,致使采集的农作物图片质量降低,严重影响了农民对田间农作物管理的判断,不利于对农作物实施精准的田间管理。以运动模糊图像为研究对象,提出对图片倒谱进行预处理,根据清晰倒谱中亮条纹方向与运动方向一致的特性,来确定模糊方向,利用其对称性来确定模糊尺度,得到相应的PSF(点扩展函数),再选用Lucy-Richardson的方法复原图像。图像恢复的主观效果表明,该方法对运动模糊图像复原效果良好,能为后续的特征提取、目标识别奠定良好的基础。
The relative movement between crop and imaging system caused the blurred motion of the image, which resulted in the decrease of the picture quality of the collected crop, which seriously affected the farmers’ judgments on the crop management in the field and not conducive to the accurate field management of the crop. Taking the motion blurred image as the research object, this paper proposes to preprocess the cepstrum of the picture, and to determine the blur direction according to the characteristics of the clear cepstrum in which the direction of the bright stripe coincides with the direction of motion. The symmetry is used to determine the fuzzy scale and the corresponding PSF (Point spread function), then use Lucy-Richardson’s method to restore the image. The subjective effects of image restoration show that this method is effective for the restoration of motion blurred images and can lay a good foundation for the subsequent feature extraction and target recognition.