Design Of Flotation Foam Image Monitoring System Based On Embedded Linux

With the development of science and technology, the progress of network techniques, electronic technology, image processing techniques and embedded technology provide the power and vitality for modern image and video monitoring system. Compared with traditional monitoring system based on personal computer, the embedded image and video monitoring system has the advantages of small power consumption,low cost and steady run. Foam flotation method is an important method of coal dressing. We can direct foam flotation production process by observing the external features of coal form. The combination of embedded monitoring system and coal slime flotation technology can avoid disadvantages of strong subjectivity based on manual monitoring and delayed operation, and can improve the automation level and production benefit of enterprise.This paper designed the coal slime foam image monitoring system based on embedded Linux. In this system, the Mini 2440(ARM9) target board played the part of hardware platform, and the ov9650 USB digital camera played the part of foam image acquisition module, and the TPO 3.5 inch TFT LCD played the part of local foam image display module. This embedded monitoring system was connected with remote personal computer monitoring client by DM9000 network adapter, and realized the monitoring for coal slime flotation process.In terms of software platform, paper used Samsung‘s Supervivi for boot loader. The boot loader, 2.6.32.6 version of Linux kernel, and system file added with Qt/Embedded graphics library were compiled, and then transplanted to the Mini2440 target board. In order to realize the function of remote monitoring through network, the network video server MJPEG-Streamer was also transplanted to this target board. On the development host, the embedded Linux cross-compiling environment was built. The mid-value filtering denoising and linear transformation function enhancement algorithm application and enhancement algorithm application by ? function parameters optimization based on UWPSO and friendly human machine interface for coal slime foam images on Mini2440 target board were accomplished. On the remote client, the 2010 a version of MATLAB was installed, and the simulation experiments about three image enhancement algorithms for coal slime foam images which were linear transformation function method, particle swarm optimization method and improved particle swarm optimization method were completed. Paper has proposed an image enhancement algorithm based on improved particle swarm optimization(PSO),and has achieved obvious enhancement effect.Paper used watershed algorithm to devide the enhanced flotation foam images,and extracted the bubble size as characteristic value for statistics and analysis.The development environment which combined OpenCV with Qt was built. Graphical design method and C++ program language were used to realized the remote monitoring and image processing human machine interface.The flotation foam image monitoring system based on embedded Linux was tested at last. The acquired foam images were processed by the enhancement algorithm based on improved PSO.And the image quality in contrast and clarity has been improved according to the test result.After the image segmentation,paper extracted the foam size feature,and combined it with flotation experiences to direct production.