The curse of dimensionality is the main reason for computational complexity and Hughes phenomenon in supervised hyperspectral classification. Previous studies seldom consider a simultaneous fashion real situation insufficiency available training samples, particularly small land covers that often contain key information scene, problem complexity. In this paper, capabilities feature reduction technique used discrimination are combined with advantages Bayesian learning-based probabilistic...

Need for a priori knowledge of the components comprising each pixel in scene has set endmember determination, rather than abundance quantification, as primary focus many unmixing approaches. In absence information about pure signatures present an image scene, which is often case, mean spectra vectors, directly extracted from are usually used signatures' spectra. This approach mathematically optimized problems with known ignores some statistical properties samples and leads to suboptimal...

An efficient superresolution technique through spatial-spectral data fusion for hyperspectral (HS) imagery is proposed in this letter. The spatial and spectral contents of an HS image are extracted using a linear mixture model fully constrained least squares unmixing technique. These then combined correlation learning-based mapping (SRM) algorithm. realistically simulates between the low-resolution (LR) its subsampled version ( LR <sup xmlns:mml="http://www.w3.org/1998/Math/MathML"...

National diagnostic reference levels (NDRLs) of Iran were determined for the four most common CT examinations including head, sinus, chest and abdomen/pelvis. A new 'quality control (QC)-based dose survey method', as developed by us, was applied to 157 scanners in (2014-15) with different slice classes, models geographic spread across country. The NDRLs abdomen/pelvis are 58, 29, 12 14 mGy CTDIVol 750, 300, 300 650 mGy.cm DLP, respectively. 'QC-based method' further proven that it is a...

Recent data suggest that knowledge of radiation exposures among physicians is inadequate. This study, therefore, aimed to evaluate their the doses patients received and awareness associated biological risks exposure. A questionnaire in multiple-choice format consisted four sections with a total 10 questions based on literature review. 136 questionnaires were returned from 69 general practitioners 67 various specialties different hospitals capital city Tehran, Iran. Fifty-four percent...

The Iran Nuclear Regulatory Authority has investigated overexposure cases in industrial radiography over a period of three years. Radiographers with thermoluminescence dosimeter (TLD) records more than 4 mSv any 2 month routine monitoring were asked to fill questionnaire explaining their points view the reasons for such relatively high doses. responses showed that 50% radiographers did not agree recorded TLD doses, although majority alternative explanations weak. main causes overexposures...

Occupational exposures in conventional and interventional radiology were investigated over a period of 10 years for all radiation workers. The statistical analysis carried out on the refined data showed that average annual effective doses 0.28 0.59 mSv measurably exposed workers 0.18 0.52 monitored 2014. More than 99.9 82.8% received less public dose limit (1 mSv) Comparing occupational levels different countries (including Iran) as well poor comparability among them. Regarding above...

A fast and efficient spatial-spectral fusion method for resolution enhancement of hyperspectral imagery is proposed in this paper. linear mixture model fully constrained least squares based unmixing algorithm are applied spectral the resulted fractional images processed using a information correlation through learning-based super-resolution mapping technique. To validate performance method, experiments carried out on real images. The obtained results reliability main advantages include its...

Abstract A new “quality-control-based (QC-based) dose survey method” has been developed for determination of diagnostic reference levels (DRL) in Computed Tomography (CT) examinations. The “QC-based is based on the use retrospective data QC documents and reports, which are typically available from National Regulatory Authority database. method was applied to 70 CT scanners Tehran, Iran, by using reports commonly used “data collection also filling each questionnaire on-site validate method....

A method for exploiting human's internal contamination data radioactive release estimation in nuclear power plant accidents is proposed. Nevertheless, such often very rough and uncertain; it accessible even toughest situations when most of the active passive monitors are damaged by accident. These can be used combination with other collectable estimating event scale severe plan accidents. The rationale behind that associated radiation workers involved early stages emergency response...

This research proposes to utilize square-loop Frequency selective surface (FSS) in a radar station increase the efficiency of communication systems and same time protect staff people against biological effects C-band microwave radiation. In station, waves propagate undetermined directions different frequency ranges which can effect on health who work indoor environment near antenna. Microwave radiation increases temperature explosive material interferes with wireless local area network...

A new spatial-spectral data fusion technique based on spectral mixture analysis and super-resolution mapping for spatial resolution enhancement of hyperspectral imagery is proposed in this paper. To end, a linear model constrained least squares unmixing algorithm are applied the resulted fractional images processed information correlation through technique. The obtained results validate effectiveness method. It doesn't need any priori scene or secondary high source data, fast.

The purpose of this study was to evaluate the effects gantry angulation on dose profiles along z-axis, CTDIW value and efficiency in head examinations. A Monte Carlo simulation model a GE LightSpeed 16-slice CT scanner developed by GATE toolkit. CTDI100 air at isocenter dosimetry phantom were measured using pencil ion chamber. Dose studied central peripheral holes thermoluminescence LiF disc dosemeters for angulations 0°, 10° 20°. Results show small shift only with increasing (maximum 24 mm...

Hyperspectral technology has introduced a new perspective in remote sensing applications but suffers from low spatial resolution. A spatial-spectral data fusion technique based on spectral mixture analysis and super-resolution mapping for resolution enhancement of hyperspectral imagery is proposed this paper. To end linear model fully constrained least squares unmixing algorithm are applied the resulted fractional images processed information correlation through technique. validate...

Despite well-developed Regulatory Authority systems for radiation protection, there are some concerns about lost-controlled and poorly controlled radioactive sources owing to abundance dispersion of the users. This survey covers current situation categories 3 through 5 (based on categorisation IAEA) in non-medical applications Iran. The investigation performed a selected group users different provinces Findings show that unauthorised supply legacy addition abandonment under regulatory...

Unrecognised pregnancy during radioisotope therapy of thyroid cancer results in hardly definable embryo/fetus exposures, particularly when the gland is already removed. Sources such difficulty include uncertainty data like commencing time, amount and distribution metastasized cells body, effect thyroidectomy on fetus dose coefficient etc. Despite all these uncertainties, estimation order most cases enough for medical legal decision-making purposes. A model adapting coefficients recommended...

Hughes phenomenon is a serious problem in supervised classification of hyperspectral images particular for small land-cover patches. A solution this through integrating the capabilities nonlinear discriminating analysis with relevance vector machine (RVM) proposed paper. It first transforms hyperdimensional data to new space better class separability. Then, multiclass RVM classifier processes transformed precise labeling classes. The results show that approach outperforms both as well...

In this paper an enhancement in target recognition remote sensing applications using artificial neural network is described. It proposed how, by a Hopfield (HNN), more accurate measures of land targets can be obtained compared with those determined the proportion image processing alone. based on applying mathematical morphology to extract candidate objects followed implementing HNN extracted features recognize object stored templates. Results suggest that useful tool for from remotely sensed imagery.

This paper discusses an accident, which occurred in one of the radiation application centres Iran, follow-up investigations as well lessons learnt. In January 2004 Regulatory Authority was informed through a university protection officer accident regarding orphan sources. Investigations revealed that Am-Be and three (137)Cs sources container were subject to extensive heat due burning melting paraffin content container; consequently, stuck side wall container, but they still undamaged no...