Software Control And Algorithm Research Of Embedded Gas Detection System

With the progress of human society and modern industrial development, various types and quantities of harmful gases in the atmosphere has increased dramatically. It is a serious threat to human health. Therefore, a harmful gas detection system design is necessary which can detect harmful gas quickly and accurately. At present, the researches of gas detection become more and more attention of scholars at home and abroad who have proposed a variety of harmful gas detection methods, of which the use of gas sensor arrays for gas detection is more extensive. In recent years, A new method of gas detection is emerging which uses porphyrin sensor arrays (PSAs) as gas sensor arrays. The method could identify gases by detection the color change information of porphyrin sensitive points in the PSA chips before and after reaction with tested gases. The method can detect harmful gases effectively which has high sensitivity and rapid detection.In this paper, a micro-trace gas detection system based on PSA and embedded ARM system. The system achieves gas detecting software Hardware integration which is designed in the embedded ARM platform. It can be convenient for rapid and accurate detection for tested gases.The paper mainly introduces software Control and PSA images processing algorithms of embedded system, of which the interactive software system is also designed. Research consists of the following:1. According to the principle and flow of gas detection, software functions needed by the system are analyzed and the design methods of the software modules are proposed.2, By analyzing the characteristics of the image PSA, the preprocessing algorithm for PSA image is proposed, which contains image clipping method, image filtering method based on morphologic, image tilt correction method based on Radon transform and power spectrum. The preprocessing algorithm for PSA images can gain a good processing effect.3, According to feature extraction requirements for the PSA images, a new automatic feature extraction algorithm is designed. A region segmentation method based on automatic mesh which generates by the combination of projection binarization and mathematical statistics is applied to separate the regions of porphyrin sensitive points automatically. The method of automatic mesh could not have absent and redundant lines. Feature values of porphyrin sensitive spots are extracted by radius ratio method which can gain characterization of porphyrin sensitive spots in the maximum extend.4. Through the analysis of the difference color features of tested gases, a recognition algorithm based on hierarchical fuzzy neural network is proposed. The algorithm divides the process of recognition algorithm for tested gases into two layers, combining with fuzzy neural network, which can not only reduce the time cost, but also improve the accuracy of recognition.5. Under the platform of integrated development environment Qt-Creator in the Linux system, interactive software system is developed. The system can achieve the functions of hardware control, PSA image acquisition, PSA image analysis, experimental data management and storage. The software system can get a good control of the gas detection process by transplanting to embedded detection equipment.