Thz Image Stripe Noise Elimination Study

The terahertz(THz) radiation occupies a large portion of the electromagnetic spectrum between the microwave and infrared bands.Since it possesses many attractively characteristic properties,THz imaging and sensing technologies can provide information not available through conventional methods such as microwave and X-ray techniques.THz waves contain rich information involved in molecular vibration and rotation,and have sub-picoseconds pulse width,low photon energy and high space-time coherence.Because its' special characteristics,THz imaging technology has huge application potentiality in biological medicine,security inspection,testing of aerospace materials aspects and so on.In the research using CW THz imaging technology to deboned in space shuttle insulating foam,the noise severely affects the future image processing.The research content is the reduction of stripe noises in THz images.The basic objective is to effectively remove stripe noises while preserving the original image details as much as possible so that the original informations could be obtained better.In this paper,we analysis the generation of the stripe noise in THz images.According analysis the characteristics for stripes,analogy noise generation,we propose two new methods for stripes denoise in frequency area,and through experiments to compare algorithm performance and evaluation.The research contents of the paper are as follows:1.Using a threshold to determine the noise's frequency points,using frequency domain filter principle to combinated threshold segmentation and frequency domain filtering,so that filtering process more intuitive,the image.2.Based on the amplitude spectrum of the image of the projection automatically determines frequency noise algorithm,than using the traditional low-pass,band stop,notch filter for image processing,statistical methods and filtering idea of combining this ways to achieve relatively simple,and better able to eliminate stripes noise.The experimental results show that both methods can effectively eliminate the stripe noise of THz images,but also can better maintain the original image information.