Automatic defect inspection system of colour filters using Taguchi-based neural network

A coloured filter is a critical part of an LCD panel, especially to present a high quality colour display. At present, the defect detection of colour filters is conducted by manual inspection in the final product stage. However, poor detection efficiency and subjective judgment of manual inspection undermine accuracy. Therefore, this study applied image processing technology and the neural network to detect surface defects of colour filters in order to prevent losses arising from incorrect detection, lower production costs, and effectively improve yield. A back-propagation neural network (BPNN) classifier was selected to train the features. The results showed that the proposed method can be successfully applied in defect detection of colour filters to reduce artificial detection errors. In addition, the Taguchi method was used with BPNN to save time searching optimal learning parameters by the trial and error method, which achieves faster convergence, smaller convergent errors and better recognition rate. The results proved that the root-mean-square error (RMSE) of the Taguchi-based BPNN at final convergence is 0.000254, and recognition rate reaches 94%. Therefore, the proposed method has good effects in detecting the micro defects of a colour filter panel.     

Sources, concentrations, and risks of naphthalene in indoor and outdoor air

Naphthalene is a ubiquitous pollutant, and very high concentrations are sometimes encountered indoors when this chemical is used as a pest repellent or deodorant. This study describes the distribution and sources of vapor-phase naphthalene concentrations in four communities in southeast Michigan, USA. Outdoors, naphthalene was measured in the communities and at a near-road site. Indoors, naphthalene levels were characterized in 288 suburban and urban homes. The median outdoor concentration was 0.15 μg/m(3), and a modest contribution from rush-hour traffic was noted. The median indoor long-term concentration was 0.89 μg/m(3), but concentrations were extremely skewed and 14% of homes exceeded 3 μg/m(3), the chronic reference concentration for non-cancer effects, 8% exceeded 10 μg/m(3), and levels reached 200 μg/m(3). The typical excess individual lifetime cancer risk was about 10(-4) and reached 10(-2) in some homes. Important sources include naphthalene's use as a pest repellent and deodorant, migration from attached garages and, to lesser extents, cigarette smoke and vehicle emissions. Excessive use as a repellent caused the highest concentrations. Naphthalene presents high risks in a subset of homes, and policies and actions to reduce exposures, for example, sales bans or restrictions, improved labeling, and consumer education, should be considered. PRACTICAL IMPLICATIONS: Long-term average concentrations of naphthalene in most homes fell into the 0.2-1.7 μg/m(3) range reported as representative in earlier studies. The highly skewed distribution of concentrations results in a subset of homes with elevated concentrations and health risks that greatly exceed US EPA and World Health Organization (WHO) guidelines. The most important indoor source is the use of naphthalene as a pest repellant or deodorant; secondary sources include presence of an attached garage, cigarette smoking, and outdoor sources. House-to-house variation was large, reflecting differences among the residences and naphthalene use practices. Stronger policies and educational efforts are needed to eliminate or modify indoor usage practices of this chemical.     

Forecasting depression in bipolar disorder

Bipolar disorder is characterized by recurrent episodes of mania and depression and affects about 1% of the adult population. The condition can have a major impact on an individual's ability to function and is associated with a long-term risk of suicide. In this paper, we report on the use of self-rated mood data to forecast the next week's depression ratings. The data used in the study have been collected using SMS text messaging and comprises one time series of approximately weekly mood ratings for each patient. We find a wide variation between series: some exhibit a large change in mean over the monitored period and there is a variation in correlation structure. Almost half of the time series are forecast better by unconditional mean than by persistence. Two methods are employed for forecasting: exponential smoothing and Gaussian process regression. Neither approach gives an improvement over a persistence baseline. We conclude that the depression time series from patients with bipolar disorder are very heterogeneous and that this constrains the accuracy of automated mood forecasting across the set of patients. However, the dataset is a valuable resource and work remains to be done that might result in clinically useful information and tools.     

Modeling spatial variations of black carbon particles in an urban highway-building environment

Highway-building environments are prevalent in metropolitan areas. This paper presents our findings in investigating pollutant transport in a highway-building environment by combing field measurement and numerical simulations. We employ and improve the Comprehensive Turbulent Aerosol Dynamics and Gas Chemistry (CTAG) model to simulate the spatial variations of black carbon (BC) concentrations near highway I-87 and an urban school in the South Bronx, New York. The results of CTAG simulations are evaluated against and agree adequately with the measurements of wind speed, wind directions, and BC concentrations. Our analysis suggests that the BC concentration at the measurement point of the urban school could decrease by 43-54% if roadside buildings were absent. Furthermore, we characterize two generalized conditions in a highway-building environment, i.e., highway-building canyon and highway viaduct-building. The former refers to the canyon between solid highway embankment and roadside buildings, where the spatial profiles of BC depend on the equivalent canyon aspect ratio and flow recirculation. The latter refers to the area between a highway viaduct (i.e., elevated highway with open space underneath) and roadside buildings, where strong flow recirculation is absent and the spatial profiles of BC are determined by the relative heights of the highway and buildings. The two configurations may occur at different locations or in the same location with different wind directions when highway geometry is complex. Our study demonstrates the importance of incorporating highway-building interaction into the assessment of human exposure to near-road air pollution. It also calls for active roles of building and highway designs in mitigating near-road exposure of urban population.     

Functional and proliferative effects of repeated low‐dose oral administration of furan in rat liver

Scope: Furan, a food contaminant formed during heat processing, induces hepatocellular tumors in rodents and high incidences of cholangiocarcinomas in rats even at the lowest dose (2 mg/kg b.w.) administered. Initial estimates suggested that human intake of furan may be as high as 3.5 μg/kg b.w./day, indicating a relatively narrow margin of exposure. The aim of this study was to establish dose–response data for cytotoxicity, regenerative cell proliferation and secondary oxidative DNA damage in livers of male F344 rats treated with furan at doses ≤2 mg/kg b.w. for 28 days. Methods and results: No significant signs of hepatotoxicity other than a mild, dose‐dependent increase in serum cholesterol and unconjugated bile acids, and no evidence of oxidative DNA damage were seen. Histopathological alterations and proliferative changes were restricted to subcapsular areas of the left and caudate liver lobes. Conclusion: Although statistically significant effects were only seen at the 2 mg/kg b.w. dose during the course of our study, a ～two and ～threefold increase in 5‐bromo‐2′‐deoxyuridine labeling index was observed at 0.1 and 0.5 mg/kg b.w., respectively, suggesting that chronic exposure to doses even below 2 mg/kg b.w. may cause proliferative changes in rat liver and highlighting the need to assess furan carcinogenicity at lower doses.