Case Adaptive Detection Models for Asbestos-Containing Building Materials Based on Hyperspectral Image Processing

This report presents guidelines for the assessment and abatement of asbestoscontaining materials in buildings based on available information. Background information is given on the history and use of asbestos-containing products in buildings, and regulations pertaining to their use. Included are control mea- sures for buildings containing asbestos materials, procedures for determining condition of the materials, and abatement techniques for containment and removal. A summary is presented of recent guide specifications and standards developed by industry, government agencies, and a standards organization which are related to asbestos-containing materials in existing buildings. These documents include guidance for the control, assessment, and abatement of such materials.

Recently in Korea, the treatment of asbestos-containing materials (ACM) in building has emerged as one of the most important environmental health issues. This study was conducted to identify the distribution and characteristics of ACM and airborne asbestos concentrations in industrial buildings in Korea. A total of 1285 presumed asbestos-containing material (PACM) samples were collected from 80 workplaces across the nation, and 40% of the PACMs contained more than 1% of asbestos. Overall, 94% of the surveyed workplaces contained ACM. The distribution of ACM did not show a significant difference by region, employment size, or industry. The total ACM area in the buildings surveyed was 436,710 m2. Ceiling tile ACM accounted for 61% (267,093 m2) of the total ACM area, followed by roof ACM (32%), surfacing ACM (6.1%), and thermal system insulation (TSI). In terms of asbestos type, 98% of total ACM was chrysotile, while crocidolite was not detected. A comparison of building material types showed that the material with the highest priority for regular management is ceiling tile, followed by roof, TSI, and surfacing material. The average airborne concentration of asbestos sampled without disturbing in-place ACM was 0.0028 fibers/cc by PCM, with all measurements below the standard of recommendation for indoor air quality in Korea (0.01 fibers/cc).

When asbestos containing materials (ACM) must be removed from the building before demolition, additional greenhouse gas (GHG) emissions are generated. However, precedent studies have not considered the removal of ACM from the building. The present study aimed to develop a model for estimating GHG emissions created by the ACM removal processes, specifically the removal of asbestos cement slates (ACS). The second objective was to use the new model to predict the total GHG emission produced by ACM removal in the entire country of Korea. First, an input-equipment inventory was established for each step of the ACS removal process. Second, an energy consumption database for each equipment type was established. Third, the total GHG emission contributed by each step of the process was calculated. The GHG emissions generated from the 1,142,688 ACS-containing buildings in Korea was estimated to total 23,778 tonCO2eq to 132,141 tonCO2eq. This study was meaningful in that the emissions generated by ACS removal have not been studied before. Furthermore, the study deals with additional problems that can be triggered by the presence of asbestos in building materials. The method provided in this study is expected to contribute greatly to the calculation of GHG emissions caused by ACM worldwide.

Risk assessment of asbestos containing materials in a deteriorated dwelling area using four different methods

7 - Mineral fibre-based building materials and their health hazards

IntroductionNearly 14 million homes were built possibly using asbestos-containing construction materials between 1950 and the mid-1980s. Certain types of asbestos were not banned until 1999. Although exposure to asbestos is...

Former employees (214) of a plant that manufactured asbestos-containing building materials (wallboard and acoustic tile) from 1958 through 1974 were screened for asbestos-related pulmonary disease. Results are presented on 182 union members whose names appeared on a 1966 seniority list in one of six departments believed to have had substantial asbestos exposure. The study population demonstrated a high frequency of pulmonary abnormalities often associated with asbestos exposure: low mean forced vital capacity percent of predicted (FVC % predicted), low mean forced expiratory volume at one second percent of predicted (FEV1 % predicted), presence of parenchymal abnormalities (40.1% with profusion scores of 1/0 or higher), and 30.5% with bilateral pleural abnormalities on chest roentgenograms. In addition, râles in two or more locations were found in 22.8% on examination of the chest, 30.6% gave a history of chronic bronchitis, and 34.8% of dyspnea grade two or higher. Calculated asbestos exposure scores, based upon participant recall, were not found to be associated with these abnormalities. The abnormal findings were not adequately explained by potential confounders such as cigarette smoking. Other notable findings in this study include the presence of smoking-adjusted decrements in pulmonary function associated with moderate profusion scores (i.e., 1/0 and 1/1) and the presence of a substantial degree of obstructive lung disease (19.2%) in this population.

Hyperspectral image processing is a very rapidly growing technology these days. It extends the scope of its applications across diversified fields such as Remote sensing, precision agriculture, medicine, Computer Vision applications and many more. Especially Hyperspectral Imagery, benefiting from its very own nature of providing rich spectral information, find many apllications in the field of Remote sensing and Earth observation. With the development of hyperspectral imaging sensor technology, the supply of HSI data has increased greatly, creating a need for proportionate rise in the HIS Processing techniques for better interpretation of data. In this context Deep Learning has become a viable option to handle huge datasets like Hyperspectral Imagery. Thus many enthusiastic researchers are showing interest in the domain of Hyperspectral Image Processing. This paper attempts to provide the basic concepts of HSI acquisition and processing at an abstract level with more focus given to Deep Learning based HSI processing tasks. So that the work will come in handy to the beginners in this field to get the preliminary concepts of HSI at a glance. To the best of our knowledge this work is first of its kind to bring to together in this study the the concepts of HSI acquisition and HSI processing.

The adsorption of volatile organic compounds by building materials reduces the pollutant concentrations in indoor air. We collected three interior building materials with adsorption potentials—latex paint, micro-carbonized plywood, and moisture-buffering siding—used the sorptive building materials test (SBMT) to determine how much they reduced indoor formaldehyde (HCHO) concentrations, and then assessed the consequent reduction in human cancer risk from HCHO inhalation. Adsorption of HCHO by building materials significantly improved the effective ventilation efficiency. For example, the equivalent ventilation rate for Celite siding—used for humidity control—was 1.44 m3/(m2·h) at 25°C, 50% relative humidity (RH); the loading factor (L) was 0.4 m2/m3, and the HCHO concentration was 0.2 ppm; this effect is equivalent to a higher ventilation rate of approximately 0.6 air changes per hour in a typical Taiwanese dwelling. There was also a substantial reduction of risk in Case MCP-2 (Cin,te: 245 μg/m3, 30°C, 50% RH): males: down 5.73 × 10−4; females: down 4.84 × 10−4). The selection of adsorptive building materials for interior surfaces, therefore, significantly reduces human inhalation of HCHO. Our findings should encourage developing and using innovative building materials that help improve indoor air quality and thus provide building occupants with healthier working and living environments.

Feature extraction (FE) is a crucial research area in hyperspectral image (HSI) processing. Recently, due to the powerful ability of deep learning (DL) to extract spatial and spectral features, DL-based FE methods have shown great potentials for HSI processing. However, most of the DL-based FE methods are supervised, and the training of them suffers from the absence of labeled samples in HSIs severely. The training issue of supervised DL-based FE methods limits their application on HSI processing. To address this issue, in this paper, a novel modified generative adversarial network (GAN) is proposed to train a DL-based feature extractor without supervision. The designed GAN consists of two components, which are a generator and a discriminator. The generator can focus on the learning of real probability distributions of data sets and the discriminator can extract spatial-spectral features with superior invariance effectively. In order to learn upsampling and downsampling strategies adaptively during FE, the proposed generator and discriminator are designed based on a fully deconvolutional subnetwork and a fully convolutional subnetwork, respectively. Moreover, a novel min-max cost function is designed for training the proposed GAN in an end-to-end fashion without supervision, by utilizing the zero-sum game relationship between the generator and discriminator. Besides, the proposed modified GAN replaces the original Jensen-Shannon divergence with the Wasserstein distance, aiming to mitigate the unstability and difficulty of the training of GAN frameworks. Experimental results on three real data sets validate the effectiveness of the proposed method.

Target detection is a key issue in processing hyperspectral images (HSIs). Spectral-identification-based algorithms are sensitive to spectral variability and noise in acquisition. In most cases, both the target spatial distributions and the spectral signatures are unknown, so each pixel is separately tested and appears as a target when it significantly differs from the background. In this paper, we propose two algorithms to improve the signal-to-noise ratio (SNR) of hyperspectral data, leading to detectors that are robust to noise. These algorithms consist in integrating adaptive spatial/spectral filtering into the adaptive matched filter and adaptive coherence estimator. Considering the HSIs as tensor data, our approach introduces a data representation involving multidimensional processing. It combines the advantages of spatial and spectral information using an alternating least square algorithm. To estimate the signal subspace dimension in each spatial mode, we extend the Akaike information criterion, and we develop an iterative algorithm for spectral-mode rank estimation. We demonstrate the interest of integrating the quadtree decomposition to perform an adaptive 3-D filtering and thereby preserve the local image characteristics. This leads to a significant improvement in terms of denoised tensor SNR and, consequently, in terms of detection probability. The performance of our method is exemplified using simulated and real-world HYperspectral Digital Imagery Collection Experiment images.

Dimensionality reduction is a key step in hyperspectral image processing. Recent investigations are looking into nonlinear manifold learning for dimensionality reduction in hyperspectral imagery. Nonlinear manifold learning methods such as Isomap, and Local Linear Embedding are use to recover the low dimensional representation of an unknown nonlinear manifold for high dimensional data where it is important to retain the neighborhood structure of the manifold. Although these algorithms use different philosophies for recovering the nonlinear manifold, they all incorporate neighborhood information from each data point to construct a weighted graph having the data points as vertices. Thus the performance of these methods highly depends on how well these neighborhoods are selected, since all subsequent steps rely on it. The k -NN algorithm is the most widely used technique for neighbor selection in manifold learning. However, it can result in a disconnected graph and it does not fully exploit spatial neighborhood information from the image in selecting points to form the neighborhoods. In this paper, recently proposed methods for constructing the weighted graph in manifold learning are studied: k -VC, k -EC and k -MST, which have the advantage of creating connected graphs and have performed well in artificial data sets. Spatial information of the hyperspectral images is included in the manifold learning process by using spatial coherence. Experiments are conducted with artificial data and hyperspectral images. For hyperspectral images, classification accuracy was used as an indirect measure to understand how the low dimensional embedding of the data reduces dimensionality while still maintaining good discrimination performance.

We present a framework for hyperspectral image (HSI) analysis validation, specifically abundance fraction estimation based on HSI measurements of water soluble dye mixtures printed on microarray chips. In our work we focus on the performance of two algorithms, the Least Absolute Shrinkage and Selection Operator (LASSO) and the Spatial LASSO (SPLASSO). The LASSO is a well known statistical method for simultaneously performing model estimation and variable selection. In the context of estimating abundance fractions in a HSI scene, the “sparse” representations provided by the LASSO are appropriate as not every pixel will be expected to contain every endmember. The SPLASSO is a novel approach we introduce here for HSI analysis which takes the framework of the LASSO algorithm a step further and incorporates the rich spatial information which is available in HSI to further improve the estimates of abundance. In our work here we introduce the dye mixture platform as a new benchmark data set for hyperspectral biomedical image processing and show our algorithm’s improvement over the standard LASSO.

Field-programmable gate arrays (FPGAs) offer promising tools for real-time on-board hyperspectral image (HSI) processing due to its higher computational speed and better reconfigurability. In this paper, the spectral matching target detection algorithm, namely adaptive coherence estimator (ACE), is implemented based on FPGA. The flow background statistics are used to calculate the background covariance matrix, and the Sherman-Morrison method is further used to process the covariance matrix inversion. The finally calculated results should separate targets from the whole pixels by a selected threshold. Evaluation of several hyperspectral data shows that the system has a higher speed than the traditional 3.60GHz CPU (about 77.74 times) without losing accuracy.

Fiber Release from Asbestos-containing Building Material by Crushing Experiments