Batch Active Learning for Multispectral and Hyperspectral Image Segmentation Using Similarity Graphs

Graph learning,when used as a semi-supervised learning(SSL)method,performs well for classification tasks with a low label rate.We provide a graph-based batch active learning pipeline for pixel/patch neighborhood multi-or hyperspectral image segmentation.Our batch active learning approach selects a collection of unlabeled pixels that satisfy a graph local maximum constraint for the active learning acquisition function that determines the relative importance of each pixel to the classification.This work builds on recent advances in the design of novel active learning acquisition functions(e.g.,the Model Change approach in arXiv:2110.07739)while adding important further developments including patch-neighborhood image analysis and batch active learning methods to further increase the accuracy and greatly increase the computational efficiency of these methods.In addition to improvements in the accuracy,our approach can greatly reduce the number of labeled pixels needed to achieve the same level of the accuracy based on randomly selected labeled pixels.

Combined process of sequencing batch reactor activated sludge process and constructed wetland for domestic wastewater treatment

By combining sequencing batch reactor (SBR) activated sludge process and constructed wetland (CW), this study is to achieve the domestic wastewater treatment. Our purpose was to determine the optimum operating parameters of the combined process. The process involved advantages and shortages of SBR and CW. Under normal temperature, the 3rd cycle (SBR’s operation cycle is 8 h: inflow for 1 h, limited aeration for 3 h, sediment for 1 h, outflow for 1 h, and idling for 2 h; CW’s hydraulic retention time (HRT) is 24.8 h and hydraulic loading is 24.5 m3/m2 d) was the best cyclic mode. The effluents can meet the standard GB/T18921-2002: "The reuse of urban recycling water: water quality standard for scenic environment use". In the 3rd cycle, the efficiency of CW was the maximum, and energy consumption of SBR was the minimum. Under the condition of low dissolved oxygen, the removing efficiency of chemical oxygen demand (COD) and ammonia was not affected obviously. Simultaneously, nitrification and denitrification phenomena occured and phosphorus was absorbed obviously.

Production of Poly-b-hydroxybutyrate by Activated Sludge in Sequencing Batch Reactor under Aerobic Conditions

In order to improve poly-β-hydroxybutyrate（PHB） production in activated sludge, the anaerobic/aerobic alternative operating sequencing batch reactor（SBR） process was applied in this paper to accumulate PHB. Effects of nutritional conditions and carbon concentration on PHB accumulation were studied. Results indicated that PHB accumulation reached the highest level and accounted for 11.2 % under anaerobic condition for phosphate limitation and 20.84 % under aerobic condition for nitrogen and phosphate limitation of mixed liquor suspended solid（MLSS）, respectively. In addition, 4 g/L was proved to be the optimum carbon concentration in both anaerobic and aerobic experiments, and the PHB accumulation reached 17.1 %（anaerobic, phosphorus limitation） and 60.4 %（aerobic, nitrogen and phosphorus limitation） of MLSS, respectively. PHB could be successfully extracted with sodium hypochlorite and chloroform method from the activated sludge. In addition, the infrared spectrum showed that the PHB sample extracted was of high purity.

Multi-label active learning by model guided distribution matching

Multi-label learning is an effective framework for learning with objects that have multiple semantic labels, and has been successfully applied into many real-world tasks, In contrast with traditional single-label learning, the cost of la- beling a multi-label example is rather high, thus it becomes an important task to train an effective multi-label learning model with as few labeled examples as possible. Active learning, which actively selects the most valuable data to query their labels, is the most important approach to reduce labeling cost. In this paper, we propose a novel approach MADM for batch mode multi-label active learning. On one hand, MADM exploits representativeness and diversity in both the feature and label space by matching the distribution between labeled and unlabeled data. On the other hand, it tends to query predicted positive instances, which are expected to be more informative than negative ones. Experiments on benchmark datasets demonstrate that the proposed approach can reduce the labeling cost significantly.