Automatic Thresholding Technique Research Articles

Automatic Extraction of Urban Road Centerlines from High-Resolution Satellite Imagery Using Automatic Thresholding and Morphological Operation Method

The commercial high-resolution imaging satellite with 1 m spatial resolution IKONOS is an important data source of information for urban planning and geographical information system (GIS) applications. In this paper, a morphological method is proposed. The proposed method combines the automatic thresholding and morphological operation techniques to extract the road centerline of the urban environment. This method intends to solve urban road centerline problems, vehicle, vegetation, building etc. Based on this morphological method, an object extractor is designed to extract road networks from highly remote sensing images. Some filters are applied in this experiment such as line reconstruction and region filling techniques to connect the disconnected road segments and remove the small redundant. Finally, the thinning algorithm is used to extract the road centerline. Experiments have been conducted on a high-resolution IKONOS and QuickBird images showing the efficiency of the proposed method.

An Enhanced Approach for Preprocessing of Mammogram Images using Inverse Daubechies Wavelet Transform and Non-Linear Diffusion

Breast cancer is the most leading cause of death in women nowadays. Screening mammography is currently the best available radiological technique for early detection of breast cancer. The detection of breast cancer is disturbed due to the existence of artifacts which reduce the rate of accuracy. For this reason, the pre-processing of mammogram images is very important in the process of breast cancer analysis because it reduces the number of false positives. This paper discusses about two existing filtering techniques and compares it with the results of a proposed filtering method. It is used to solve the noise removal problems and separate the background region from the breast profile region using an automatic thresholding technique. The results are evaluated on the pre-processing method on a set of images obtained from MIAS database. Thus this preparation phase improves the image quality and accentuates the CAD results more accurately. General Terms Data Mining, Image mining, Breast Cancer Detection, Mammograms.

Self-adaptive partial discharge signal de-noising based on ensemble empirical mode decomposition and automatic morphological thresholding

This paper proposes a self-adaptive technique for partial discharge (PD) signal denoising with automatic threshold determination based on ensemble empirical mode decomposition (EEMD) and mathematical morphology. By introducing extra noise in the decomposition process, EEMD can effectively separate the original signal into different intrinsic mode functions (IMFs) with distinctive frequency scales. Through the kurtosis-based selection criterion, the IMFs embedded with PD impulses can be extracted for reconstruction. On the basis of mathematical morphology, an automatic morphological thresholding (AMT) technique is developed to form upper and lower thresholds for automatically eliminating the residual noise while maintaining the PD signals. The results on both simulated and real PD signals show that the above PD denoising technique is superior to wavelet transform (WT) and conventional EMD-based PD de-noising techniques.

Self-adaptive partial discharge signal de-noising based on ensemble empirical mode decomposition and automatic morphological thresholding

This paper proposes a self-adaptive technique for partial discharge (PD) signal denoising with automatic threshold determination based on ensemble empirical mode decomposition (EEMD) and mathematical morphology. By introducing extra noise in the decomposition process, EEMD can effectively separate the original signal into different intrinsic mode functions (IMFs) with distinctive frequency scales. Through the kurtosis-based selection criterion, the IMFs embedded with PD impulses can be extracted for reconstruction. On the basis of mathematical morphology, an automatic morphological thresholding (AMT) technique is developed to form upper and lower thresholds for automatically eliminating the residual noise while maintaining the PD signals. The results on both simulated and real PD signals show that the above PD denoising technique is superior to wavelet transform (WT) and conventional EMD-based PD de-noising techniques.

Biometric recognition using fuzzy score level fusion

Multi-modal biometric recognition uses more than one biometric identifier to recognise a person. Identification based on multiple templates becomes an emerging need. Multi-modal biometric systems are expected to be more reliable due to the presence of multiple independent pieces of biometric traits. This paper proposes a new approach for biometric recognition using hand vein and finger vein images. Our proposed method employs hyper analytic wavelet transform and automatic thresholding techniques to extract features from the images of hand. Once the veins are extracted the location and width are stored in the database. Here score level fusion based on fuzzy logic is utilised for recognition. This approach is tested on the finger vein and hand vein databases which is containing 500 samples from 100 users. The experimental results exhibits that the proposed method is comparable with the existing recognising methods.

A hybrid plaque characterization method using intravascular ultrasound images

Intravascular ultrasound (IVUS) is an invasive imaging modality that provides high resolution cross-sectional images permitting detailed evaluation of the lumen, outer vessel wall and plaque morphology and evaluation of its composition. Over the last years several methodologies have been proposed which allow automated processing of the IVUS data and reliable segmentation of the regions of interest or characterization of the type of the plaque. In this paper we present a novel methodology for the automated identification of different plaque components in grayscale IVUS images. The proposed method is based on a hybrid approach that incorporates both image processing techniques and classification algorithms and allows classification of the plaque into three different categories: Hard Calcified, Hard-Non Calcified and Soft plaque. Annotations by two experts on 8 IVUS examinations were used to train and test our method. The combination of an automatic thresholding technique and active contours coupled with a Random Forest classifier provided reliable results with an overall classification accuracy of 86.14%. The proposed method can accurately detect the plaque using grayscale IVUS images and can be used to assess plaque composition for both clinical and research purposes.

White Matter Hyperintensities in Mild Lewy Body Dementia

Background: The objective of this study was to explore the load of white matter hyperintensities (WMH) in patients with Lewy body dementia (LBD) and compare to Alzheimer’s disease (AD) and normal controls (NC). Methods: Diagnosis of LBD and AD was made according to consensus criteria and cognitive tests were administered. MRI scans for 77 (61 AD and 16 LBD) patients and 37 healthy elderly control subjects were available for analysis. We segmented WMH from FLAIR images using an automatic thresholding technique and calculated the volume of WMH in several regions of the brain, using non-parametric tests to compare groups. Multivariate regression was applied. Results: There were no significant differences in WMH between AD and LBD. We found a significant correlation between total and frontal WMH and Mini-Mental State Examination (MMSE) and verbal fluency scores in the AD group, but not in the LBD group. Conclusion: The WMH load in LBD was similar to that of AD. A correlation between WMH load and cognition was found in the AD group, but not in the LBD group, suggesting that vascular disease contributes to cognitive impairment in AD but not LBD.

Digital image sensor-based assessment of the status of oat (Avena sativa L.) crops after frost damage.

The aim of this paper is to classify the land covered with oat crops, and the quantification of frost damage on oats, while plants are still in the flowering stage. The images are taken by a digital colour camera CCD-based sensor. Unsupervised classification methods are applied because the plants present different spectral signatures, depending on two main factors: illumination and the affected state. The colour space used in this application is CIELab, based on the decomposition of the colour in three channels, because it is the closest to human colour perception. The histogram of each channel is successively split into regions by thresholding. The best threshold to be applied is automatically obtained as a combination of three thresholding strategies: (a) Otsu’s method, (b) Isodata algorithm, and (c) Fuzzy thresholding. The fusion of these automatic thresholding techniques and the design of the classification strategy are some of the main findings of the paper, which allows an estimation of the damages and a prediction of the oat production.

English

Image segmentation, the division of a multi-dimensional image into groups of associated pixels, is an essential step for many advanced imaging applications. Image segmentation can be performed by recursively splitting the whole image or by merging together a large number of minute regions until a specified condition is satisfied. The split-and-merge procedure of image segmentation takes an intermediate level in an image description as the starting cutest, and thereby achieves a compromise between merging small primitive regions and recursively splitting the whole images to reach the desired final cutest. The proposed segmentation approach is a split-andmerge technique. The conventional split-and-merge algorithm is lacking in adaptability to the image semantics because of its stiff quadtree-based structure. In this paper, an automatic thresholding technique based on bimodality detection approach with non-homogeneity criterion is employed in the splitting phase of the split-and-merge segmentation scheme to directly reflect the image semantics to the image segmentation results. Since the proposed splitting technique depends upon homogeneity factor, some of the split regions may or may not split properly. There should be rechecking through merging technique between the two adjacent regions to overcome the drawback of the splitting technique. A sequential-arrange-based or a minimal spanning-tree based approach, that depends on data dimensionality of the weighted centroids of all split regions for finding the pair wise adjacent regions, is introduced. Finally, to overcome the problems caused by the splitting technique, a novel merging technique based on the density ratio of the adjacent pair regions is proposed. The algorithm has been tested on several synthetic as well as real life data and the results show the efficiency of the segmentation technique. Defence Science Journal, 2010, 60(3), pp.290-301 , DOI:http://dx.doi.org/10.14429/dsj.60.356

Structural Invariance of Sunspot Umbrae over the Solar Cycle: 1993 – 2004

Measurements of maximum magnetic flux, minimum intensity, and size are presented for 12 967 sunspot umbrae detected on the National Aeronautics and Space Administration/National Solar Observatory (NASA/NSO) spectromagnetograms between 1993 and 2004 to study umbral structure and strength during the solar cycle. The umbrae are selected using an automated thresholding technique. Measured umbral intensities are first corrected for center-to-limb intensity dependence. Log-normal fits to the observed size distribution confirm that the size-spectrum shape does not vary with time. The intensity – magnetic-flux relationship is found to be steady over the solar cycle. The dependence of umbral size on the magnetic flux and minimum intensity are also independent of the cycle phase and give linear and quadratic relations, respectively. While the large sample size does show a low-amplitude oscillation in the mean minimum intensity and maximum magnetic flux correlated with the solar cycle, this can be explained in terms of variations in the mean umbral size. These size variations, however, are small and do not substantiate a meaningful change in the size spectrum of the umbrae generated by the Sun. Thus, in contrast to previous reports, the observations suggest the equilibrium structure, as manifested by the invariant size-magnetic field relationship, as well as the mean size (i.e., strength) of sunspot umbrae do not significantly depend on the solar-cycle phase.

Computer-aided Detection of Ischemic Lesions Related to Subcortical Vascular Dementia on Magnetic Resonance Images

The purpose of this study was to develop an automated method for detection of the hyperintense ischemic lesions related to subcortical vascular dementia based on conventional magnetic resonance images (T1-weighted, T2-weighted, and fluid-attenuated inversion-recovery images [FLAIR]). Our proposed method was based on subtraction between the T1-weighted image and the FLAIR image. First, a brain region was extracted by an automated thresholding technique based on a linear discriminant analysis for a pixel value histogram. Next, for enhancement of ischemic lesions, the T1-weighted image was subtracted from the fluid-attenuated inversion-recovery image. Ischemic lesion candidates were identified using a multiple gray-level thresholding technique and a feature-based region-growing technique on the subtraction image. Finally, an artificial neural network trained with 15 image features of the ischemic candidates was used to remove false-positives. We applied our method to nine patients with vascular dementia (age range, 64-94 years, mean age, 69.4 years; four males and five females), who were scanned on a 1.5-T magnetic resonance unit. Our method achieved a sensitivity of 90% with 4.0 false-positives per slice in detection of ischemic lesions. The overlap measure between ischemic lesion areas obtained by our method and a neuroradiologist was 60.7% on average. The ratio of ischemic lesion area to the whole brain area obtained by our method correlated with that determined by a neuroradiologist with a correlation coefficient of 0.911. Our preliminary results suggest that the proposed method may have feasibility for evaluation of the ischemic lesion area ratio.

Segmentation of defects from radiography images by the histogram concavity threshold method

A histogram concavity-based thresholding approach has been used for segmenting porosities, voids and inclusions from digitised radiography images. Studies on small defects in cylindrical tubes and flat plates show that histogram concavity-based thresholding approach yields better results compared to conventional automatic thresholding techniques like Maximum Entropy and Otsu methods, where heterogeneous background present in the image due to the geometry and large X-ray quantum and structural noise gives poor segmentation. The segmentation procedure is optimised by filtering the abrupt noises with low-pass median filtering, followed by grey-level global thresholding. The histogram concavity-based thresholds provided excellent segmentation of pores and voids present in Electrochemical Hydrogen Sensors (ECHS), cast plates used in refineries, micro-pores in tube-to-tubesheet (TTS) welds of steam generator of fast breeder reactor, tungsten inclusion in Tungsten Inert Gas (TIG) welds, crater crack and external undercut in welds. An accuracy of ± 1-2 pixel sizes (65-130 microns) is obtained for the estimation of defects sizes from the grey-level profiling.

Remote Sensing Change Detection Based on Canonical Correlation Analysis and Contextual Bayes Decision

In this paper, we present a new approach combining canonical correlation analysis and contextual Bayes decision for change detection in bi-temporal multispectral remotely sensed images. Canonical correlation analysis in the form of Minimum Noise Fraction/Multivariate Alteration Detection transformation was first applied to two multispectral images to effectively yield a difference image, followed by a contextual Bayes decision procedure using automatic thresholding and Markov random field modeling techniques to identify areas where changes may have actually occurred from the difference image. An experiment of monitoring land reclamations in Hong Kong from Landsat TM/ETM+ images was conducted, and the results demonstrated effectiveness of our approach.

Generalized minimum-error thresholding for unsupervised change detection from SAR amplitude imagery

The availability of synthetic aperture radar (SAR) data offers great potential for environmental monitoring due to the insensitiveness of SAR imagery to atmospheric and sunlight-illumination conditions. In addition, the short revisit time provided by future SAR-based missions will allow a huge amount of multitemporal SAR data to become systematically available for monitoring applications. In this paper, the problem of detecting the changes that occurred on the ground by analyzing SAR imagery is addressed by a completely unsupervised approach, i.e., by developing an automatic thresholding technique. The image-ratioing approach to SAR change detection is adopted, and the Kittler and Illingworth minimum-error thresholding algorithm is generalized to take into account the non-Gaussian distribution of the amplitude values of SAR images. In particular, a SAR-specific parametric modeling approach for the ratio image is proposed and integrated into the thresholding process. Experimental results, which confirm the accuracy of the method for real X-band SAR and spaceborne imaging radar C-band images, are presented

Assessment of automatic gap fraction estimation of forests from digital hemispherical photography

Thresholding is a central part of the analysis of hemispherical images in terms of gap fraction and leaf area index (LAI), and the selection of optimal thresholds has remained a challenge over decades. The need for an objective, automatic, operator-independent thresholding method has long been of interest to scientists using hemispherical photography. This manuscript deals with the comparison of a wide variety of different well-known automatic thresholding techniques against the subjective manual method, using high-dynamic range digital hemispherical photographs. The performance of the different thresholding methods was evaluated based on: (1) visual inspection by means of a multi-criteria decision system and (2) quantitative analysis of the methods’ sensitivity to an overall performance criterium. The automatic Ridler clustering method proved to be the most robust thresholding method for various canopy structure conditions. This automatic method might be the best solution for a fast, reliable and objective use of hemispherical photographs for gap fraction and LAI estimation in forest stands, given that the threshold setting is no longer manually performed. The fine-tuning potential of local thresholding methods to better address particular photographic limitations (e.g. over-exposure in a certain image region) is also presented.

A novel watershed method using second order gradient image

Thresholding is an important step in the segmentation of image features, and the existing methods are not all effective when the image histogram exhibits a unimodal pattern, which is common in defect detection of fruit. This study was aimed at developing a general automatic thresholding methodology for fast and effective segmentation of bruises from the images acquired by structured-illumination reflectance imaging (SIRI). SIRI images, under sinusoidal patterns of illumination at a spatial frequency of 100 cycles m−1, were acquired from 120 apple samples of four varieties with artificially created bruises and from another 40 apples with naturally occurred bruises. Subsequently, three sets of images, i.e., amplitude component (AC), direct component (DC) and ratio (i.e., dividing AC by DC), were derived from the original SIRI images. A unimodal thresholding method, called UNIMODE, was first applied to DC images for background removal, and then nine automatic thresholding techniques, including one unimodal and eight bimodal, were applied to the ratio images for bruise segmentation. It was found that severe over-segmentation occurred when using the bimodal thresholding methods, and this problem was mitigated by confining threshold selection to the lower part of the histogram that contained bruise information. Three bimodal thresholding techniques, i.e., INTERMODE (histogram valley emphasized), RIDLER (iterative thresholding), OTSU (clustering based) achieved the best bruise detection results with the overall accuracies of more than 90%. The overall detection results were further improved by integrating these techniques with the unimodal thresholding, due to reductions in the false positive error. The three bimodal thresholding techniques resulted in overall detection accuracies of 77–85% for naturally occurred bruises. This study has showed that the proposed automatic thresholding methodology provides a simple and effective tool for bruise detection of apples.

An evaluation of global thresholding techniques for the automatic image segmentation of automotive aluminum sheet alloys

Digital image analysis is a valuable tool for the quantification of microstructural morphology, such as the size and clustering distributions of second-phase particles. However, the first step to such analysis is the segmentation of digital micrographs to identify the features of interest. Since the outcome of this step will have a significant influence on all subsequent processing, accurate segmentation is of the utmost importance. Grayscale images of the microstructures of three automotive aluminum alloys are used (AA5182, AA5754, and AA6111) to compare the abilities of several automatic global image thresholding techniques. The minimum cross-entropy method of Li and Lee [Pattern Recognit. 30 (1993) 617] was found to work well for all alloys when combined with the cost function of Yen et al. [IEEE T. Image Process. 4 (1995) 370].

Real-time thresholding with Euler numbers

The problem of finding an automatic thresholding technique is well known in applications involving image differencing like visual-based surveillance systems, autonomous vehicle driving, etc. Among the algorithms proposed in the past years, the thresholding technique based on the stable Euler number method is considered one of the most promising in terms of visual results. Unfortunately its high computational complexity made it an impossible choice for real-time applications. The implementation here proposed, called fast Euler numbers, overcomes the problem since it calculates all the Euler numbers in just one single raster scan of the image. That is, it runs in O(hw), where h and w are the image’s height and width, respectively. A technique for determining the optimal threshold, called zero crossing, is also proposed.

A MAHALANOBIS DISTANCE BASED CLASSIFIER FOR DIAGNOSIS OF DISEASES

ABSTRACT Multi-dimensional information, including many different biochemical testing items, is conventionally used to characterize medical examination data. However, ambiguous and inconsistent multi-dimensional data are difficult to explore the relationship among them. The conventional means of developing an expert system requires formulating rules to analyze the input data. Formulating such rules is quite difficult with large sets of input data. This study presents a classification approach for diagnosing medical examination information. This is a multivariate analysis approach based on Mahalanobis distance (MD) classifier, creates the MD space in advance using the homogenous examination data. Then, an automatic thresholding technique adopted as a decision criterion in a MD space, is based on the maximum variance between-classes. Comparing the MD classification approach with the neural network approach reveals that the MD classifier is highly effective in diagnosing liver diseases because it has a classification accuracy exceeding 95%. In addition, the MD classifier outperforms neural network classifier by increasing the correct classification up to 5.97% for the testing sets. Finally, some characteristics are discussed.

Diffusion-perfusion MR evaluation of perihematomal injury in hyperacute intracerebral hemorrhage

It has been suggested that a zone of perihematomal ischemia analogous to an ischemic penumbra exists in patients with primary intracerebral hemorrhage (ICH). Diffusion-perfusion MRI provides a novel means of assessing injury in perihematomal regions in patients with ICH. To characterize diffusion-perfusion MRI changes in the perihematomal region in patients with hyperacute intracerebral hemorrhage. Twelve patients presenting with hyperacute, primary ICH undergoing CT scanning and diffusion-perfusion MRI within 6 hours of symptom onset were reviewed. An automated thresholding technique was used to identify decreased apparent diffusion coefficient (ADC) values in the perihematomal regions. Perfusion maps were examined for regions of relative hypo- or hyperperfusion. Median baseline NIH Stroke Scale score was 17 (range, 6 to 28). Median hematoma volume was 13.3 mL (range, 3.0 to 74.8 mL). MRI detected the hematoma in all patients on echo-planar susceptibility-weighted imaging and in all seven patients imaged with gradient echo sequences. In six patients who underwent perfusion imaging, no focal defects were visualized on perfusion maps in tissues adjacent to the hematoma; however, five of six patients demonstrated diffuse ipsilateral hemispheric hypoperfusion. On diffusion imaging, perihematomal regions of decreased ADC values were identified in three of 12 patients. All three subsequently showed clinical and radiologic deterioration. A rim of perihematomal decreased ADC values was visualized in the hyperacute period in a subset of patients with ICH. The presence of a rim of decreased ADC outside the hematoma correlated with poor clinical outcome. Although perfusion maps did not demonstrate a focal zone of perihematomal decreased blood flow in any patient, most patients had ipsilateral hemispheric hypoperfusion.