Food recognition using neural network classifier and multiple hypotheses image segmentation

ABSTRACT

This paper proposes the multiple-hypotheses image segmentation and feed-forward neural network classifier for food recognition to improve the performance. Initially, the food or meal image is given as input. Then, the segmentation is applied to identify the regions, where a particular food item is located using salient region detection, multi-scale segmentation, and fast rejection. Then, the features of every food item are extracted by the global feature and local feature extraction. After the features are obtained, the classification is performed for each segmented region using a feed-forward neural network model. Finally, the calorie value is computed with the aid of (i) food volume and (ii) calorie and nutrition measure based on mass value. The experimental results and performance evaluation are validated. The outcome of the proposed method attains 0.947 for Macro Average Accuracy (MAA) and 0.959 for Standard Accuracy (SA), which provides better classification performance.     

Demonstration of a tool for automatic learning and re-use of knowledge in the activated sludge process.

Wastewater treatment plant operators encounter complex operational problems related to the activated sludge process and usually respond to these by applying their own intuition and by taking advantage of what they have learnt from past experiences of similar problems. However, previous process experiences are not easy to integrate in numerical control, and new tools must be developed to enable re-use of plant operating experience. The aim of this paper is to investigate the usefulness of a case-based reasoning (CBR) approach to apply learning and re-use of knowledge gained during past incidents to confront actual complex problems through the IWA/COST Benchmark protocol. A case study shows that the proposed CBR system achieves a significant improvement of the benchmark plant performance when facing a high-flow event disturbance.     

Ecophysiology and niche differentiation of Nitrospira-like bacteria, the key nitrite oxidizers in wastewater treatment plants.

Nitrite-oxidizing bacteria of the genus Nitrospira are key nitrifiers in wastewater treatment plants. Pure cultures of these organisms are unavailable, but cultivation-independent molecular methods make it possible to detect Nitrospira-like bacteria in environmental samples and to investigate their ecophysiology. Comprehensive screening of natural and engineered habitats and of public databases for 16S rRNA sequences of Nitrospira-like bacteria revealed a surprisingly high biodiversity in the genus Nitrospira, which comprises at least four phylogenetic sublineages. All Nitrospira-like bacteria detected in wastewater treatment plants belonged to the sublineages I and II. Subsequently, the population dynamics of different Nitrospira-like bacteria were monitored, by quantitative fluorescence in situ hybridization with rRNA-targeted probes, confocal laser scanning microscopy and digital image analysis, during incubation of nitrifying activated sludge in media containing different nitrite concentrations. These experiments showed that Nitrospira-like bacteria, which were affiliated with the phylogenetic sublineages I or II of the genus Nitrospira, responded differently to nitrite concentration shifts. Previously unknown properties of Nitrospira-like bacteria were discovered in the course of an environmental genomics project. Implications of the obtained results for fundamental understanding of the microbial ecology of nitrite oxidizers as well as for future improvement of nutrient removal in wastewater treatment plants are discussed.     

Upflow anaerobic sludge blanket (UASB) treatment of supernatant of cow manure by thermal pre-treatment.

Upflow anaerobic sludge blanket (UASB) methane fermentation treatment of cow manure that was subjected to screw pressing, thermal treatment and subsequent solid-liquid separation was studied. Conducting batch scale tests at temperatures between 140 and 180 degrees C, the optimal temperature for sludge settling and the color suppression was found to be between 160-170 degrees C. UASB treatment was carried out with a supernatant obtained from the thermal treatment at the optimal conditions (170 degrees C for 30 minutes) and polymer-dosed solid-liquid separation. In the UASB treatment with a COD(Cr) loading of 11.7 kg/m3/d and water temperature of 32.2 degrees C, the COD(Cr) level dropped from 16,360 mg/L in raw water to 3,940 mg/L in treated water (COD(Cr), removal rate of 75.9%), and the methane production rate per COD(Cr) was 0.187 Nm3/kg. Using wastewater thermal-treated at the optimal conditions, also a methane fermentation treatment with a continuously stirred tank reactor (CSTR) was conducted (COD(Cr) in raw water: 38,000 mg/L, hydraulic retention time (HRT): 20 days, 35 degrees C). At the COD(Cr) loading of 1.9 kg/m3/d, the methane production rate per COD(Cr), was 0.153 Nm3/kg. This result shows that UASB treatment using thermal pre-treatment provides a COD(Cr), loading of four times or more and a methane production rate of 1.3 times higher than the CSTR treatment.