Global Thresholding Techniques Research Articles

Automated Defective Ceramic Tiles Classification using Image Processing Techniques

Surface quality is one of the critical issues in the ceramic tile manufacturing industry. However, there are still large amounts of ceramic tiles manufacturers who use a manual inspection process to check the quality of the ceramic tiles’ surfaces. The manual inspection is time-consuming and has low accuracy. Additionally, the slow inspection process could not keep up with the production rate. Therefore, this study presents a solution where image processing techniques are implemented to construct an automated defective ceramic tiles classification. The process consists of a few phases which include image enhancement, image segmentation, feature extraction, and classification. The input image which is the ceramic tiles image was enhanced using the histogram equalization technique. The segmentation was done using Sobel and the global thresholding technique before the feature extraction process is implemented. Three shape features of mean intensity, area, and perimeter were extracted in analyzing the characteristics of each ceramic tile defect type of crack, corner, and edge. Then, the image was classified using the K-Nearest Neighbor (k-NN) technique. A confusion matrix was used to assess the performance of the k-NN classification to 150 testing images. The overall mean accuracy, sensitivity, and specificity returned good performance at 93.34%, 89.93%, and 95% respectively. Thus, it can be inferred that the proposed automated defective ceramic tiles classification using image processing techniques was successful.

Estimation of SPIO Nanoparticles Uptakes by Macrophages Using Transmission Electron Microscopy.

Due to their interesting size-dependent magnetic characteristics and relative biocompatibility, magnetic superparamagnetic iron oxide (SPIO) nanoparticles have been widely exploited as probes for cell and subcellular structure identification, as well as medication and gene delivery. A thorough understanding of the mechanics of the interaction between nanoparticles and macrophages is vital in managing dynamic processes in nanomedicine. In this study, the interaction behavior and uptake of SPIO nanoparticles by M1- and M2-type macrophages were investigated. Mice monocytes were differentiated into M1 and M2 macrophages, and the uptake of SPIO nanoparticles was studied using a TEM microscope. A high resolution image of 1 nm resolution, an image processing technique, was developed to extract the SPIO-NPs from tomographic TEM microscopic images. Lysosomes appear to be the zones of high concentrations of SPIO inside macrophages. Lysosomes were first selected in each image, and then segmentation by the Otsu thresholding method was used to extract the SPIO-NPs. The Otsu threshold method is a global thresholding technique used to automatically differentiate SPIOs from the background. The SPIO-NPs appear in red colors, and the other pixels in the image are considered background. Then, an estimation of the SPIO-NP uptakes by lysosomes is produced. Higher uptake of all-sized nanoparticles was observed in M1- and M2-type macrophages. An accurate estimation of the number of SPIO-NPs was obtained. This result will help in controlling targeted drug delivery and assessing the safety impact of the use of SPIO-NPs in nanomedicine for humans.

A CISG Method for Internal Defect Detection of Solar Cells in Different Production Processes

The internal defect detection of solar cells indifferent production processes currently adopts manual visual verification on the images captured by electroluminescence or photoluminescence system. To improve the efficiency and reliability of the inspection, this article proposes a generic and automatic component-of-interest superposition graph(CISG) method. First, the solar cell inspection region is located by shape-based matching. Second, based on the properties of Fourier transform and singular value decomposition, the frequency components of the testing image are modified adaptively, and then component graphs with different local saliency are generated. According to the normalized marginal gains of energy, these component graphs are combined into a saliency map. Finally, defective regions are extracted from the saliency map using the global threshold technique. By employing various compared approaches to conduct experiments on five datasets from different production processes, the great performance of our presented method is revealed, which also implies this method outperforms the current methods.

Fractional Frequency Reuse Optimal SINR Threshold Selection Using NIR and ISODATA

A vital part of cellular network evolution has been long-term evolution networks. In these networks, it is important to mitigate inter-cell interference. Fractional frequency re-use has been proposed to address this. The method involves the division of cells into two regions based on a signal-to-interference-plus-noise-ratio threshold value. The inner region adopts a frequency re-use of one (1), while the outer region uses a higher frequency re-use factor. Setting the threshold value is a critical problem addressed in this paper. The proposed approach adapts techniques used in image processing called global-thresholding techniques. The approaches considered are iterative self-organizing data analysis and native integral ratio. Mobile stations in a cell continuously report their signal-to-interference-plus-noise-ratio values to the base station. These reported values are used to determine a threshold which dictates which subscribers fall in the inner and outer regions. The threshold value is periodically updated based on the new reported values over time. Simulations are used to assess the performance using throughput and fairness metrics. By setting the threshold optimally, better throughputs and fairness are then achieved. We concluded that native integral ratio marginally outperformed the iterative self-organizing data analysis method, and it significantly outperformed static fractional frequency reuse techniques.

Multilevel Thresholding for Image Segmentation Using Mean Gradient

Image binarization and segmentation have been one of the most important operations in digital image processing and related fields. In spite of the enormous number of research studies in this field over the years, huge challenges still exist hampering the usability of some existing algorithms. Some of these challenges include high computational cost, insufficient performance, lack of generalization and flexibility, lack of capacity to capture various image degradations, and many more. These challenges present difficulties in the choice of the algorithm to use, and sometimes, it is practically impossible to implement these algorithms in a low-capacity hardware application where computational power and memory utilization are of great concern. In this study, a simple yet effective and noniterative global and bilevel thresholding technique is proposed. It uses the concept of image gradient vector to binarize or segment the image into three clusters. In addition, a parametric preprocessing approach is also proposed that can be used in image restoration applications. Evidences from the experiments from both visual and standard evaluation metrics show that the proposed methods perform exceptionally well. The proposed global thresholding outperforms the formidable Otsu thresholding technique.

Adaptive Digital Hologram Binarization Method Based on Local Thresholding, Block Division and Error Diffusion.

High-speed optical reconstruction of 3D-scenes can be achieved using digital holography with binary digital micromirror devices (DMD) or a ferroelectric spatial light modulator (fSLM). There are many algorithms for binarizing digital holograms. The most common are methods based on global and local thresholding and error diffusion techniques. In addition, hologram binarization is used in optical encryption, data compression, beam shaping, 3D-displays, nanofabrication, materials characterization, etc. This paper proposes an adaptive binarization method based on a combination of local threshold processing, hologram division into blocks, and error diffusion procedure (the LDE method). The method is applied for binarization of optically recorded and computer-generated digital holograms of flat objects and three-dimensional scenes. The quality of reconstructed images was compared with different methods of error diffusion and thresholding. Image reconstruction quality was up to 22% higher by various metrics than that one for standard binarization methods. The optical hologram reconstruction using DMD confirms the results of the numerical simulations.

Automated Detection of Brain Tumor through Magnetic Resonance Images Using Convolutional Neural Network.

Brain tumor is a fatal disease, caused by the growth of abnormal cells in the brain tissues. Therefore, early and accurate detection of this disease can save patient's life. This paper proposes a novel framework for the detection of brain tumor using magnetic resonance (MR) images. The framework is based on the fully convolutional neural network (FCNN) and transfer learning techniques. The proposed framework has five stages which are preprocessing, skull stripping, CNN-based tumor segmentation, postprocessing, and transfer learning-based brain tumor binary classification. In preprocessing, the MR images are filtered to eliminate the noise and are improve the contrast. For segmentation of brain tumor images, the proposed CNN architecture is used, and for postprocessing, the global threshold technique is utilized to eliminate small nontumor regions that enhanced segmentation results. In classification, GoogleNet model is employed on three publicly available datasets. The experimental results depict that the proposed method is achieved average accuracies of 96.50%, 97.50%, and 98% for segmentation and 96.49%, 97.31%, and 98.79% for classification of brain tumor on BRATS2018, BRATS2019, and BRATS2020 datasets, respectively. The outcomes demonstrate that the proposed framework is effective and efficient that attained high performance on BRATS2020 dataset than the other two datasets. According to the experimentation results, the proposed framework outperforms other recent studies in the literature. In addition, this research will uphold doctors and clinicians for automatic diagnosis of brain tumor disease.

Quadtree partitioning scheme of color image based

Image segmentation is an essential complementary process in digital image processing and computer vision, but mostly utilizes simple segmentation techniques, such as fixed partitioning scheme and global thresholding techniques due to their simplicity and popularity, in spite of their inefficiency. This paper introduces a new split-merge segmentation process for a quadtree scheme of colour images, based on exploiting the spatial and spectral information embedded within the bands and between bands, respectively. The results show that this technique is efficient in terms of quality of segmentation and time, which can be used in standard techniques as alternative to a fixed partitioning scheme.

Quadtree partitioning scheme of color image based

Image segmentation is an essential complementary process in digital image processing and computer vision, but mostly utilizes simple segmentation techniques, such as fixed partitioning scheme and global thresholding techniques due to their simplicity and popularity, in spite of their inefficiency. This paper introduces a new split-merge segmentation process for a quadtree scheme of colour images, based on exploiting the spatial and spectral information embedded within the bands and between bands, respectively. The results show that this technique is efficient in terms of quality of segmentation and time, which can be used in standard techniques as alternative to a fixed partitioning scheme.

Aluminium Alloy Foam Modelling and Prediction of Elastic Properties Using X-ray Microcomputed Tomography

X-ray microcomputed tomography has been gaining relevance in the field of cellular materials to characterize materials and analyse their microstructure. So, here, it was used together with finite element modelling to develop numerical models to estimate the effective properties (Young’s modulus) of aluminium alloy foams and evaluate the effects of processing on the results. A manual global thresholding technique using the mass as a quality indicator was used. The models were reconstructed (Marching Cubes 33), then simplified and analysed in terms of mass and shape maintenance (Hausdorff distance algorithm) and face quality. Two simplification procedures were evaluated, with and without small structural imperfections, to evaluate the impact of the procedures on the results. Results demonstrate that the developed procedures are good at minimizing changes in mass and shape of the geometries while providing good face quality, i.e., face aspect ratio. The models are also shown to be able to predict the effective properties of metallic foams in accordance with the findings of other researchers. In addition, the process of obtaining the models and the presence of small structural imperfections were shown to have a great impact on the results.

A novel computer aided diagnostic system for quantification of metabolites in brain cancer

Cancer diagnosis is one of the challenging tasks in medical image processing. First time, we have explored a novel method that locates tumors and its sub-regions using magnetic resonance images (MRI) and magnetic resonance spectroscopy (MRS) image processing. The structural based features of cancerous cells were estimated using MRI images and MRS spectra to evaluate the quantitative analysis of metabolites. An automated computer aided diagnosis (CAD) system was developed based on morphological operations and Otsu’s Global Thresholding (OGT) technique which efficiently differentiates the normal and abnormal cells of brain. In addition, multi thresholding technique was applied to delineate intertumoral and peritumoral areas such as outer edema, necrosis, and malignancy. Brain cancer tissues were categorized into eight groups such as astrocytomas, meningiomas, lymphoma, medulloblastoma, gliomas, neoplasm, metastasis, and abscess. The proposed CAD system extracts tumor regions by inspecting them with fast and interactive morphological selection-based region analysis techniques. The segmentation of cancerous tissue was compared with MRSI to assess and extract the abnormal tissue region using textural based features. A comprehensive analysis of metabolites ratio presented the high diagnostic accuracy in high- and low-grade tumor. Experimental results of the proposed CAD system achieved 100% sensitivity, 94.73% specificity and 97.14% accuracy performance parameters in brain tumors classification which demonstrated the efficacy of cancer cells segmentation with metabolites spectra. The performance of the proposed CAD system was also compared with radiologists reports in order to use as medical diagnostic tool.

Implementation of online and offline product selection system using FCNN deep learning: Product analysis

Now a day’s Artificial intelligence and deep learning techniques recommended critical E-commerce applications. The human computing, computer aided designs cannot understand the alternative offline and online products. Therefore, customers are critical to find out the products such as groceries, fashion and health. However, it is a major task to overcome this limitation for human perceptions. In this research work an advanced FCNN deep learning model is proposed with global thresholding technique. For this work selecting the digital images and online images for pre-processing and classification. At primary stage segmentation is applied then classification is performed through FCNN, at final calculating the performance measures such as accuracy 98.7%, sensitivity 98.7% and throughput 99.23% has been achieved and outcomes are challenging the present technology.

Automatic Segmentation of Retinal Capillaries in Adaptive Optics Scanning Laser Ophthalmoscope Perfusion Images Using a Convolutional Neural Network.

PurposeAdaptive optics scanning laser ophthalmoscope (AOSLO) capillary perfusion images can possess large variations in contrast, intensity, and background signal, thereby limiting the use of global or adaptive thresholding techniques for automatic segmentation. We sought to develop an automated approach to segment perfused capillaries in AOSLO images.Methods12,979 image patches were extracted from manually segmented AOSLO montages from 14 eyes and used to train a convolutional neural network (CNN) that classified pixels as capillaries, large vessels, background, or image canvas. 1764 patches were extracted from AOSLO montages of four separate subjects, and were segmented manually by two raters (ground truth) and automatically by the CNN, an Otsu's approach, and a Frangi approach. A modified Dice coefficient was created to account for slight spatial differences between the same manually and CNN-segmented capillaries.ResultsCNN capillary segmentation had an accuracy (0.94), a Dice coefficient (0.67), and a modified Dice coefficient (0.90) that were significantly higher than other automated approaches (P < 0.05). There were no significant differences in capillary density and mean segment length between manual ground-truth and CNN segmentations (P > 0.05).ConclusionsClose agreement between the CNN and manual segmentations enables robust and objective quantification of perfused capillary metrics. The developed CNN is time and computationally efficient, and distinguishes capillaries from areas containing diffuse background signal and larger underlying vessels.Translational RelevanceThis automatic segmentation algorithm greatly increases the efficiency of quantifying AOSLO capillary perfusion images.

Global and Adaptive Thresholding Technique for White Blood Cell Image Segmentation

Global and local thresholding are two thresholding approaches for white blood cell (WBC) image segmentation. Global thresholding determines the threshold value based on the histogram of the overall pixel intensity distribution of the image. In contrast, adaptive thresholding computes the threshold value for each fractional region of the image, so that each fractional region has a different threshold value. In this work, we are assessing both of these approaches for two threshold values. We extended the Otsu’s equation to calculate more than one threshold as it originally designed to find only a single threshold value. Adaptive thresholding first divides an image into fractional-image by considering an imaginary bounding box that surrounds the location of WBC, which involves the Gram-Schmidt orthogonalization method. For segmentation performance evaluation, we compare 35 blood smear test images which segmented by our proposed method, with their corresponding ground truth image to representing them in Zijdenbos Similarity Index (ZSI), precision, and recall measurement. Experimental results show that adaptive thresholding achieves average ZSI, precision and recall, 92.5%, 91.79%, 94.03%, while global thresholding achieves 30.72%, 23.38%, and 99.39% respectively.

Automatic Salient Object Extraction Based on Locally Adaptive Thresholding to Generate Tactile Graphics

Automatic extraction of salient regions is beneficial for various computer vision applications, such as image segmentation and object recognition. The salient visual information across images is very useful and plays a significant role for the visually impaired in identifying tactile information. In this paper, we introduce a novel saliency cuts method using local adaptive thresholding to obtain four regions from a given saliency map. First, we produced four regions for image segmentation using a saliency map as an input image and local adaptive thresholding. Second, the four regions were used to initialize an iterative version of the GrabCuts algorithm and to produce a robust and high-quality binary mask with a full resolution. Finally, salient objects’ outer boundaries and inner edges were detected using the solution from our previous research. Experimental results showed that local adaptive thresholding using integral images can produce a more robust binary mask compared to the results from previous works that make use of global thresholding techniques for salient object segmentation. The proposed method can extract salient objects with a low-quality saliency map, achieving a promising performance compared to existing methods. The proposed method has advantages in extracting salient objects and generating simple, important edges from natural scene images efficiently for delivering visually salient information to the visually impaired.

Extraction of Vasculature Map of Color Retinal Fundus Image

This paper presents an unsupervised and novel method for extraction of vasculature map from colored retinal fundus images. Proposed technique makes use of a fusion of bimodal masking and global thresholding technique for extraction of vessels. For this, adaptive histogram equalization method is used for enhancement of the retinal input fundus images while, on the other hand, an average filter is used on the masked images to remove the noise from the image. At this stage, bimodal masking is used to generate the masked image to exclude the pixels that belong to the background. The use of this technique reduces the analysis time and computational effort as operations will be focused only on the object pixels. After that, a global thresholding technique is utilized on the masked foreground image, which produces a vasculature map with border. Since the border has no concern in our system, hence, it is removed using the mask generated through bimodal masking. Fundus images of DRIVE and STARE database is used to perform extensive computations. The results are encouraging as the proposed system shows better outcomes on 3-quality measures: average sensitivity, specificity, and accuracy which comes out to be (84.18%, 96.68% and 95.68%) and (81.79%, 90.74% and 90.08%) for DRIVE and STARE database respectively. The results prove that the proposed methodology is capable of extracting the vasculature map accurately, which can be further helpful for diagnosis, screening, and treatment of various disorders of fundus images.

Medical Image Processing of Thermal Images in Light of Applied Color Palettes

The choice of colour palette in image processing of medical thermal images plays a crucial role and it essentially affects the interpretation and visualization of the thermal image. Even though it has remained a subject of criticism due to the difference of visualization as compared to the grayscale colour palette, the rainbow palette has been widely applied in medical image processing of thermal images. In this work, the rainbow colour palette and the grayscale colour palette were applied on the medical thermal images representing skin conditions and a comparative study was performed to assess their applicability and performance in the methodology applied. It is observed that both the palettes are suitable at different stages as part of methodology to study medical thermal images. Using the global thresholding techniques for foreground detection, the grayscale palette was found useful for image segmentation into background and foreground while the rainbow palette is effective when colour-based clustering is performed to segregate the object of interest into varying temperature zones.

Exploration and evaluation of efficient pre-processing and segmentation technique for breast cancer diagnosis based on mammograms

Breast cancer is the second leading cause of death for women everywhere in the world. Since the reason behind the disease remains unknown, early detection and diagnosis is the key challenge for breast cancer control. In this work, mammogram images are initially subject to pre-processing using Laplacian filter for enhancement of tumour regions, Gaussian mixture model, Gaussian kernel FCM, Otsu global thresholding and FCM technique are employed for segmentation. Further, the efficiency of segmentation techniques is analyzed by classifying the samples into benign, malignant and healthy using Gray Level Co-occurrence Matrix (GLCM) features. Linear discriminant analysis classifier is used a combination based on which efficiency used for classification of mammograms. Ensemble methods are evaluated. The efficiency has resulted in better accuracy with the ensemble-based method. The experimentation is conducted in the mini MIAS database of mammograms, and the efficiency of the linear discriminant analyzer is found to be 89.19% for GKFCM, 83.78% with Otsu and 78.38% with FCM method with GLCM features.

Comparative appraisal of global and local thresholding methods for binarisation of off-axis digital holograms

Binary digital and computer-generated holograms are used in a number of digital micromirror device (DMD) applications, including holographic displays, media characterisation, optical encryption, and others. Binarisation is one of the most simple and efficient methods of hologram compression. In this paper, 12 local and 18 global thresholding techniques of different groups were analysed and compared based on attribute similarity, clustering, entropy, histogram shape, and local adaptive factors. Optically recorded off-axis digital holograms of different objects were binarised using these methods. The amplitude quality of the obtained reconstructed images was compared using PSNR and SSIM values. The clustering-based methods achieved the highest quality. The results of binarisation by global and local methods were comparable on average.

Porosity Determination of Carbon and Glass Fibre Reinforced Polymers Using Phase-Contrast Imaging

This paper presents multi-modal image data of different fibre reinforced polymer samples acquired with a desktop Talbot-Lau grating interferometer (TLGI) X-ray computed tomography (XCT) system and compare the results with images acquired using conventional absorption-based XCT. Two different fibre reinforced polymer samples are investigated: (i) a carbon fibre reinforced polymer (CFRP) featuring a copper mesh embedded near the surface for lightning conduction and (ii) a short glass fibre reinforced polymer (GFRP) sample. The primary goal is the non-destructive detection of internal defects such as pores and the quantification of porosity. TLGI provides three imaging modalities including attenuation contrast (AC) due to absorption, differential phase contrast (DPC) due to refraction and dark-field contrast (DFC) due to scattering. In the case of the CFRP sample, DPC is less prone to metal streak artefacts improving the detection of pores that are located close to metal components. In addition, results of a metal artefact reduction (MAR) method, based on sinogram inpainting and an image fusion concept for AC, DPC and DPC, are presented. In the case of the GFRP sample, DPC between glass fibres and matrix is lower compared to AC while DPC shows an increased contrast between pores and its matrix. Porosity for the CFRP sample is determined by applying an appropriate global thresholding technique while an additional background removal is necessary for the GFRP sample.