Po Kim Lo

Local adaptive approach toward segmentation of microscopic images of activated sludge flocs

Abstract. Activated sludge process is a widely used method to treat domestic and industrial effluents. The conditions of activated sludge wastewater treatment plant (AS-WWTP) are related to the morphological properties of flocs (microbial aggregates) and filaments, and are required to be monitored for normal operation of the plant. Image processing and analysis is a potential time-efficient monitoring tool for AS-WWTPs. Local adaptive segmentation algorithms are proposed for bright-field microscopic images of activated sludge flocs. Two basic modules are suggested for Otsu thresholding-based local adaptive algorithms with irregular illumination compensation. The performance of the algorithms has been compared with state-of-the-art local adaptive algorithms of Sauvola, Bradley, Feng, and c-mean. The comparisons are done using a number of region- and nonregion-based metrics at different microscopic magnifications and quantification of flocs. The performance metrics show that the proposed algorithms performed better and, in some cases, were comparable to the state-of the-art algorithms. The performance metrics were also assessed subjectively for their suitability for segmentations of activated sludge images. The region-based metrics such as false negative ratio, sensitivity, and negative predictive value gave inconsistent results as compared to other segmentation assessment metrics.

Illumination Compensated Segmentation of Microscopic Images of Activated Sludge Flocs

Image processing and analysis is a useful tool for monitoring of activated sludge wastewater treatment plant. However its effectiveness is dependent on performance of the segmentation algorithms. The activated sludge plant is monitored by image processing and analysis of images acquired through trinocular microscope. The sample observed under microscope is collected from aeration tank of the plant. In this paper, a segmentation technique with integrated illumination compensation is proposed for the microscopic images of the activated sludge samples. The illumination noise was modeled and estimated as Gaussian distribution symmetric about a threshold value determined by global Otsu thresholding algorithm. The performance of the algorithm was evaluated using time required for segmentation, Rand index, accuracy and quantification of flocs. In order to compare with the state-of-the-art algorithms, gold approximations of ground truth images were manually prepared. The performance was assessed by combining the evaluation metrics in an integrated perspective. The proposed algorithm exhibits better performance in terms of both integrated and non-integrated perspectives.

Generalized classification modeling of activated sludge process based on microscopic image analysis

ABSTRACT The state of activated sludge wastewater treatment process (AS WWTP) is conventionally identified by physico-chemical measurements which are costly, time-consuming and have associated environmental hazards. Image processing and analysis-based linear regression modeling has been used to monitor the AS WWTP. But it is plant- and state-specific in the sense that it cannot be generalized to multiple plants and states. Generalized classification modeling for state identification is the main objective of this work. By generalized classification, we mean that the identification model does not require any prior information about the state of the plant, and the resultant identification is valid for any plant in any state. In this paper, the generalized classification model for the AS process is proposed based on features extracted using morphological parameters of flocs. The images of the AS samples, collected from aeration tanks of nine plants, are acquired through bright-field microscopy. Feature-selection is performed in context of classification using sequential feature selection and least absolute shrinkage and selection operator. A support vector machine (SVM)-based state identification strategy was proposed with a new agreement solver module for imbalanced data of the states of AS plants. The classification results were compared with state-of-the-art multiclass SVMs (one-vs.-one and one-vs.-all), and ensemble classifiers using the performance metrics: accuracy, recall, specificity, precision, F measure and kappa coefficient (κ). The proposed strategy exhibits better results by identification of different states of different plants with accuracy 0.9423, and κ 0.6681 for the minority class data of bulking.

Investigation on the performance of hybrid anaerobic membrane bioreactors for fouling control and biogas production in palm oil mill effluent treatment

Three different sizes of powdered activated carbon (PAC) were added in hybrid anaerobic membrane bioreactors (AnMBRs) and their performance was compared with a conventional AnMBR without PAC in treating palm oil mill effluent. Their working volume was 1 L each. From the result, AnMBRs with PAC performed better than the AnMBR without PAC. It was also found that adding a relatively smaller size of PAC (approximately 100 μm) enhanced the chemical oxygen demand removal efficiency to 78.53 ± 0.66%, while the concentration of mixed liquor suspended solid and mixed liquor volatile suspended solid were 8,050 and 6,850 mg/L, respectively. The smaller size of PAC could also enhance the biofloc formation and biogas production. In addition, the smaller particle sizes of PAC incorporated into polyethersulfone membrane resulted in higher performance of membrane fouling control and produced better quality of effluent as compared to the membrane without the addition of PAC.

Performance of duckweed and effective microbes in reducing arsenic in paddy and paddy soil

In this study phytoremediation plant (duckweed) and effective microbes were used to investigate their effectiveness in reducing arsenic concentration in paddy soil and paddy grain. The results show that using duckweed alone is a better choice as it could decrease the arsenic concentration in paddy by 27.697 % and 8.268 % in paddy grain and paddy husk respectively. The study also found out that the concentration of arsenic in soil would affect the performance of duckweed and also delayed the reproduction rate of duckweed. Using the mixture of effective microbes and duckweed together to decrease arsenic in paddy was noticed having the least potential in reducing the arsenic concentration in paddy.In this study phytoremediation plant (duckweed) and effective microbes were used to investigate their effectiveness in reducing arsenic concentration in paddy soil and paddy grain. The results show that using duckweed alone is a better choice as it could decrease the arsenic concentration in paddy by 27.697 % and 8.268 % in paddy grain and paddy husk respectively. The study also found out that the concentration of arsenic in soil would affect the performance of duckweed and also delayed the reproduction rate of duckweed. Using the mixture of effective microbes and duckweed together to decrease arsenic in paddy was noticed having the least potential in reducing the arsenic concentration in paddy.

Estimation of sludge volume index (SVI) using bright field activated sludge images

Image processing and analysis is a potential tool for monitoring of activated sludge wastewater treatment plant. One of the important parameters to track the performance of activated sludge plant is sludge volume index (SVI). In this paper, image analysis based modeling is used to estimate the sludge volume index. Bright field microscopic images of activated sludge were segmented by integrating four algorithms to skim any possible failures of any of them. The morphological parameters for activated sludge flocs have been extracted from the segmented images. Seven classes were identified for image analysis parameters with respect to range of equivalent diameter of activated sludge flocs. The process resulted into 134 image analysis parameters and seven classes. The feature selection is done by two procedures: correlation method and stepwise linear regression. The stepwise linear regression is automated process which selected 6 parameters with adjusted correlation of 95.1%. The results showed that image analysis based modeling with as small as six parameters can be used to predict the sludge volume index. Finally, three out of seven classes are identified which can contribute to the estimation of SVI.

Segmentation of activated sludge flocs with modeling of illumination noise

Fault diagnosis of activated sludge wastewater treatment plant for abnormal operation can be done using image processing and analysis of microscopic images of samples collected from aeration tank of the plant. In this paper, a novel illumination compensated segmentation technique is proposed for bright field microscopic images of the activated sludge wastewater samples. The illumination noise is modeled as Gaussian distribution and used with global Otsu thresholding. The performance of the algorithm is assessed using accuracy and Rand index. The segmentation is assessed using gold approximations of ground truth images, which were prepared manually. The proposed algorithm is compared with the local adaptive algorithms of Sauvola and Bradley. The performance metrics showed better performance of the proposed algorithm.

Iterative region based Otsu thresholding of bright-field microscopic images of activated sludge

Image processing and analysis is a potential alternative monitoring tool for biodegradation in activated sludge wastewater treatment process. Accuracy of image analysis based predictive models depends on the quality of the segmentation of the microbial aggregates in microscopic images of activated sludge. The segmentation of the images is hindered by irregular illumination and properties of the microbial aggregates such as varying opaqueness and size. In this paper, an iterative region based Otsu thresholding is proposed for the bright field microscopic images of activated sludge. The suggested approach takes not only the statistics of grayscale intensities into account but also the regional distribution of the illumination noise. The proposed algorithm is compared with state-of-the-art Otsu, iterative Otsu and local Otsu segmentation techniques. The performance of the algorithms is assessed using accuracy, Rand index (RI) and variation of information (VI). The proposed algorithm exhibited better performance in terms of all the metrics with accuracy 0.9854, RI 0.9721 and VI 0.2141.

A Vignetting Correction Algorithm for Bright-Field Microscopic Images of Activated Sludge

Vignetting correction is an essential part of pre-processing to address decreasing illumination towards the border of a microscopic image. Image processing and analysis is a potential tool for monitoring and prediction of activated sludge wastewater treatment plant. Microscopic images of activated sludge require anti-vignetting or vignetting correction algorithm for accuracy of subsequent image analysis procedures. In this paper, we proposed a vignetting correction procedure based on Gaussian modeling. The Gaussian model is estimated by using Otsu threshold and iterative evaluation of the model. The advantage of the proposed algorithm is that the vignetting model estimated for one image is found to be valid for all other images irrespective of illumination of microscope. Once the model is calibrated, the correction becomes simple addition, making the proposed technique time-efficient compared to state-of-the-art procedures. The assessment was done subjectively and by using segmentation. The proposed procedure performed better than polynomial approximation and comparable to Leongs algorithm.