Research On Compression Algorithm Of Ultrasonic Testing Data

Data compression is one of the important functions of ultrasonic equipment,which can relieve the pressure of storage space.In this paper,ultrasonic data acquisition is carried out on the 32x64 phased array ultrasonic testing.The generation process of ultrasonic data is studied,and to reduce the information entropy of the data by preprocessing.Futhermore,the compression algorithms suitable for ultrasonic data are studied,and compression solutions of different platforms are given.The main work and research results are as follows:(1)Feature research and waveform separation of ultrasonic data.The characteristics of ultrasonic defect signal are studied,and a waveform classification method based on threshold value is proposed,which divides ultrasonic A-scan data into defect and grass data.(2)Ultrasonic data preprocessing.According to the characteristics of grass and defect wave,the appropriate pretreatment method is adopted respectively.The grass data is random,so smooth preprocessing is used.The defect wave data presents a continuous change,so the differential preprocessing is carried out.(3)The research of lossless compression algorithm.Combined with the characteristics of grass wave,the RLE is modified to reduce the compression rate.The real-time compression speed of Huffman encoding is improved by using the pipeline working mode.At the same time,the ultrasonic data is compressed by using DEFLATE algorithm,and a good compression performance is obtained.(4)The research of lossy compression algorithm.Wavelet transform compression of ultrasonic A-sacn data is carried out by using db3 wavelet base function and hard threshold quantization,and its compression rate is 14.58%.Using multimedia technology to compress ultrasonic data,and obtain a good compression effect.(5)Testing and analysis.The lossy and lossless compression algorithms are tested on PC platform and embedded platform respectively.The experimental results show that the compression effect of DEFLATE,a lossless compression algorithm,and MPEG,a lossy compression algorithm,are better than others in PC platform,and the compression rate are 6.8% and 1.5% respectively.In the embedded platform,in order to meet the requirements of realtime compression,this paper presents an algorithm combination,DEFLATE compresses defect data and RLE compresses grass data,and compression rate reached 10%.