Automatic Detection And Contour Extraction Of Microorganisms In Sewage

Recent studies have confirmed that the statistical properties of microorganisms in sewage,such as their population,abundance and activities play a key indicative role in quality judgement of wastewater treatment.Automatic identification and quantitative analysis by using image processing technology will be helpful for wastewater treatment and energy saving optimization control,and will promote the development of microbial detection as well.Therefore,it has a wonderful scientific research and application value.This paper focuses on the researches in terms of the indicative microbe automatic detection and the target contour extraction,which are based on the microscopic video images from urban sewage treatment process with activated sludge.The main work and contributions are as follows:1.In this paper,a new moving object detection algorithm based on codebook background subtract is proposed.Firstly,we build the background model using the pixels in first frame and their random adjacent points to efficiently reduce the modeling time.Secondly,a new color space and a fast local directional pattern texture operator have been introduced into this model to increase the target contrast and neighborhood dependency.Finally,a self-adaptive model updating strategy based on statistical parameter estimation has been proposed for updating the codebook model quickly and effectively.Experimental results have demonstrated that the proposed method can effectively resolve such problems as body deformation,complicated background and multi-target interference et al.2.We present a novel 3D self-organizing neural network moving object detection algorithm.Firstly,we design a multilayer network topology instead of the traditional single-layer self-organizing map(SOM),which significantly improves the discrimination ability of moving objects.Secondly,new background model initialization and adaptively update mechanisms weaken the bootstrapping and ghost influences.Thirdly,we create buffer layers in neural network efficiently resolve the dynamic background and variable motion problems.Finally,a simple Kalman predictor with constant coefficients has been constructed to tackle with the cases of microorganism being obscured or lost.The experimental results show that the proposed method can effectively resolve such difficulties as variable motion,dynamic background and data loss et al.3.A microbial contour extraction algorithm based on edge connection is put forward in this study.Firstly,we use the sum of gradient direction to preliminary determine the edge regions which can suppress the high intensity noise.Then,an adaptive threshold system is designed to extract the real edges effectively.Next,combining the gradient amplitude,gradient direction difference and the characteristics of grey value,we can accurately determine the edge points excluding the noise resistance.Finally,an edge model introduced to realize the reasonable edges connection.Compared with other traditional approaches based on edge detection,our proposed method achieves more complete and smooth microbe contours.4.In this paper,we also present a neural network microbe contour extraction method based on SOM.Firstly,we use SOM to cluster the gradient of image to avoid gradient threshold human selection.Then,matching time model is introduced to avoid the influence of sample order on network convergence.Finally,according to the result of clustering,the algorithm can automatically determine the network layer numbers and realize weak edge and contour extraction by self-organizing learning.