Number Of Pixels Research Articles

The Development of Various Methods for Object Tracking and Classification in Thermal Videos

Nowadays, thermal cameras play a major role because of its temperature-based photography in many applications such as video surveillance, monitoring electronics/electrical machines, forest monitoring, monitoring babies/adult patients, and suspicious object detection. Tracking pedestrians in thermal video is a major task for such applications. Thermal cameras usually create images based on temperature emitted by the object only and not on the lighting conditions and outdoor environment conditions. But still thermal images have constraints like no texture or colour information, more number of dead pixels, low resolution, and noticeable visual colour patterns in case of any temperature variations. So the challenge in tracking pedestrians in thermal videos is tracking objects/pedestrians throughout the video without an identity switch by overcoming these constraints which may mislead the tracking process. To overcome these constraints, the proposed system uses tri feature matrix (TFM) as an object descriptor which is used to uniquely identify and represent objects in thermal images. TFM is represented in more compact way as a triple matrix. It is a simple and accurate descriptor suitable for tracking objects in thermal video sequences without an identity switch. The proposed Pedestrian tracking system uses most of the advantages of thermal cameras by overcoming challenges in thermal videos effectively based on a novel descriptor TFM. The proposed system is evaluated with various data sets, and the results are analysed using true positive, true negative, false negative, false positive, accuracy, precision, recall, F-score, global identity mismatch (GMME) and track matching error (TME). The performance metrics such as accuracy, precision, recall, F-score, GMME and TME are computed as 99%, 100%, 99%, 99%, 2.3%, and 2.1%, respectively. From the observation, it is found that the performance of proposed TFM-based system is significantly improved. The experimental result shows that the proposed system achieved more accurate tracking compared to the conventional methods

A forensic framework to improve digital image evidence administration in IIoT✰

The extensive growth of Industrial Internet of Things (IIoT) has prompted cyber attackers to commence a variety of attacks which need to be identified to maintain the security of industrial data and services. Digital forensics is an extravagant need to detect cyber attacks and to identify cybercriminals. While conducting criminal investigation using digital forensics, there is high probability of leakage of collected digital evidence. Hence, the need for a secure and lightweight cryptosystem arises to protect it from cyber-attacks. In the present paper, authors aim to design a novel forensic framework for digital evidence administration using advanced encryption and preservation mechanisms. Authors focus to encrypt and preserve only digital image evidence but can be deployed on other types of multimedia data like video frames. Three types of Chaos-map algorithms namely 4D hyperchaotic map, 2D Arnold Cat map (ACM) and Henon Map have been utilized to encrypty and decrypt digital image evidence which enable high sensitivity and dependency on initial conditions, quasi-randomness, and ergodicity to digital content. The proposed approaches have been analyzed on different groups of images of different size and validated using various security parameters such as unified average changing intensity (UACI), correlation coefficient, and the number of pixels changing rate (NPCR). Image transformation and compression has been performed after encryption process using Fast Fourier Transformation (FFT). The proposed work has been analyzed using histogram and auto-correlation. Experimentations have provided the values of the correlation coefficient of plain images near 1 and the value of adjacent cipher images near zero which show a strong correlation. The evaluated results for UACI are more than 33 % for every considered images. The value for NPCR should be near 100 % and here the calculated results for each image are above 99 %. The complexity of the proposed encryption algorithm is O(n) due to key generation using a cryptographically secure pseudo-random number generator (CSPRNG). The compressed and transformed image evidence have been stored into decentralized blockchain to ensure the integrity and confidentiality. Decentralized blockchain has been used as a reliable and promising solution to preserve digital evidence due to its distinct features like secure peer to peer communication, improved trust and transparency among participants, integrity, confidentiality, immutability, without risk of single point-of-failure. The proposed digital forensic model is able to prevent various threats and security challenges which can contaminate or tamper off digital pieces of evidence and assures improved digital evidence administration and defensibility. In the proposed scheme, the encryption process takes only 2–3 s to perform the encryption of each size of image.

Reference-free in situ rapid regional calibration of phase-only spatial light modulators.

Spatial light modulators (SLMs) have become an indispensable element in modern optics for their versatile performance in many applications. Among various types of SLMs, such as digital micromirror devices (DMD), liquid crystal-based phase-only spatial light modulators (LC-SLMs), and deformable mirrors (DM), LC-SLMs are often the method of choice due to their high efficiency, precise phase modulation, and abundant number of effective pixels. In general, for research grade applications, an additional SLM calibration step is required due to fabrication imperfection resulting in non-flat liquid crystal panels and varying phase responses over the SLM area. Here, we demonstrate a straightforward approach for reference-free orthogonal calibration of an arbitrary number of SLM subregions which only requires the same measurement time as global calibration. The proposed method requires minimal optical elements and can be applied to any optical setup as is. As a benchmark performance test, we achieved a 2.2-fold enhancement in correction efficiency for wavefront shaping through scattering media utilizing the calibrated 2160 subregions of the SLM, in comparison with a single global look-up table (LUT).

Differential perovskite hemispherical photodetector for intelligent imaging and location tracking

Advanced photodetectors with intelligent functions are expected to take an important role in future technology. However, completing complex detection tasks within a limited number of pixels is still challenging. Here, we report a differential perovskite hemispherical photodetector serving as a smart locator for intelligent imaging and location tracking. The high external quantum efficiency (~1000%) and low noise (10−13 A Hz−0.5) of perovskite hemispherical photodetector enable stable and large variations in signal response. Analysing the differential light response of only 8 pixels with the computer algorithm can realize the capability of colorful imaging and a computational spectral resolution of 4.7 nm in a low-cost and lensless device geometry. Through machine learning to mimic the differential current signal under different applied biases, one more dimensional detection information can be recorded, for dynamically tracking the running trajectory of an object in a three-dimensional space or two-dimensional plane with a color classification function.

Beyond Conventional Monitoring: A Semantic Segmentation Approach to Quantifying Traffic-Induced Dust on Unsealed Roads.

Road dust is a mixture of fine and coarse particles released into the air due to an external force, such as tire-ground friction or wind, which is harmful to human health when inhaled. Continuous dust emission from the road surfaces is detrimental to the road itself and the road users. Due to this, multiple dust monitoring and control techniques are currently adopted in the world. The current dust monitoring methods require expensive equipment and expertise. This study introduces a novel pragmatic and robust approach to quantifying traffic-induced road dust using a deep learning method called semantic segmentation. Based on the authors' previous works, the best-performing semantic segmentation machine learning models were selected and used to identify dust in an image pixel-wise. The total number of dust pixels was then correlated with real-world dust measurements obtained from a research-grade dust monitor. Our method shows that semantic segmentation can be adopted to quantify traffic-induced dust reasonably. Over 90% of the predictions from both correlations fall in true positive quadrant, indicating that when dust concentrations are below the threshold, the segmentation can accurately predict them. The results were validated and extended for real-time application. Our code implementation is publicly available.

Guidewire Endpoint Detection Based on Pixel‐Adjacent Relation during Robot‐Assisted Intravascular Catheterization: In Vivo Mammalian Models

Existing surgical guidewire endpoint localization methods in X‐ray images face challenges owing to their small size, simple appearance, nonrigid nature of objects, low signal‐to‐noise ratio of X‐ray images, and imbalance between the number of guidewire and background pixels, which lead to errors in surgical navigation. An eight‐neighborhood‐based method for increasing the localization accuracy of guidewire endpoint to improve the safety of interventional procedures is proposed herein. The proposed method includes two stages: 1) An improved U‐Net network is employed for segmenting the data of the guidewire to extract regions of interest containing guidewire endpoints with higher precision and to reduce interference from other anatomical structures and imaging artifacts. 2) The proposed method detects guidewire endpoints using the adjacent relationship between pixels in the eight‐neighborhood regions. This stage covers skeletonization extraction, removal of bifurcation points, and repair of fracture points. This study achieves mean pixel errors of 2.02 and 2.13 pixels in an in vivo rabbit and porcine X‐ray fluoroscopy images, outperforming ten classic heatmap and regression methods, achieving state‐of‐the‐art detection results. The proposed method can also be applied to detect other tiny surgical instruments such as stents and balloons, while preserving the flexibility of the guidewire bending angle.

A forward-modelling approach to overcome point spread function smearing and fit flexible models to the chemical structure of galaxies

ABSTRACT Historically, metallicity profiles of galaxies have been modelled using a radially symmetric, two-parameter linear model, which reveals that most galaxies are more metal-rich in their central regions than their outskirts. However, this model is known to yield inaccurate results when the point spread function (PSF) of a telescope is large. Furthermore, a radially symmetric model cannot capture asymmetric structures within a galaxy. In this work, we present an extension of the popular forward-modelling Python package lenstronomy, which allows the user to overcome both of these obstacles. We demonstrate the new features of this code base through two illustrative examples on simulated data. First, we show that through forward modelling, lenstronomy is able to recover accurately the metallicity gradients of galaxies, even when the PSF is comparable with the size of a galaxy, as long as the data are observed with a sufficient number of pixels. Additionally, we demonstrate how lenstronomy is able to fit irregular metallicity profiles to galaxies that are not well-described by a simple surface brightness profile. This opens up pathways for detailed investigations into the connections between morphology and chemical structure for galaxies at cosmological distances using the transformative capabilities of JWST. Our code is publicly available and open source, and can also be used to model spatial distributions of other galaxy properties that are traced by its surface brightness profile.

Spatial-Temporal resolution implementation of cloud-aerosols data through satellite cross-correlation

The Moderate Resolution Imaging Spectroradiometer (MODIS) instrument aboard Terra and Aqua satellites provides measurements of several atmospheric parameters.This paper focuses on the cloud fraction data representing the number of cloudy pixels divided by the total number of pixels, and available through 1° x 1° grids spatial resolution with daily or monthly temporal resolution. The aim of the study is to propose a novel method called The Spatial-Temporal Implementation Algorithm (STIA) for analysing satellite daily 1° x 1°grid cloud fraction average values for•Comparing two datasets retrieved by MODIS aboard Aqua and Terra satellites to obtain information on the cloud formation in the afternoon and morning, respectively, thus enhancing the temporal resolution.•Comparing the actual parameter with others retrieved by instruments aboard of different satellites characterized by a better resolution. As an example of STIA application, this study uses the Aerosol Optical Depth (AOD) collected by the Ozone Monitoring Instrument (OMI) on board of Aura satellite for comparison with MODIS instrument to achieve and enhanced spatial resolution of the grid-cell.

Dynamic light scattering and laser speckle contrast imaging of the brain: theory of the spatial and temporal statistics of speckle pattern evolution.

Dynamic light scattering (DLS) and laser speckle contrast imaging (LSCI) are closely related techniques that exploit the statistics of speckle patterns, which can be utilized to measure cerebral blood flow (CBF). Conventionally, the temporal speckle intensity auto-correlation function is calculated in DLS, while the spatial speckle contrast Ks is calculated in LSCI measurements. Due to the rapid development of CMOS detection technology with increased camera frame rates while still maintaining a large number of pixels, the ensemble or spatial average of as well as the temporal contrast Kt can be easily calculated and utilized to quantify CBF. Although many models have been established, a proper summary is still lacking to fully characterize DLS and LSCI measurements for spatial and temporal statistics, laser coherence properties, various motion types, etc. As a result, there are many instances where theoretical models are misused. For instance, mathematical formulas derived in the diffusive regime or for ergodic systems are sometimes applied to small animal brain measurements, e.g., mice brains, where the assumptions are not valid. Therefore, we aim to provide a review of the speckle theory for both DLS and LSCI measurements with detailed derivations from first principles, taking into account non-ergodicity, spatial and temporal statistics of speckles, scatterer motion types, and laser coherence properties. From these calculations, we elaborate on the differences between spatial and temporal averaging for DLS and LSCI measurements that are typically ignored but can result in inaccurate measurements of blood flow, particularly the spatially varying nature of the static component in and Kt. We also obtained maps in in vivo mouse brain measurements using high frame rate CMOS cameras which have not been demonstrated before, and compared with and Ks,t. This work provides a useful guide for choosing the correct model to analyze spatial and temporal speckle statistics in in-vivo DLS and LSCI measurements.

Scan-less microscopy based on acousto-optic encoded illumination.

Several optical microscopy methods are now available for characterizing scientific and industrial processes at sub-micron resolution. However, they are often ill-suited for imaging rapid events. Limited by the trade-off between camera frame-rate and sensitivity, or the need for mechanical scanning, current microscopes are optimized for imaging at hundreds of frames-per-second (fps), well-below what is needed in processes such as neuronal signaling or moving parts in manufacturing lines. Here, we present a scan-less technology that allows sub-micrometric imaging at thousands of fps. It is based on combining a single-pixel camera with parallelized encoded illumination. We use two acousto-optic deflectors (AODs) placed in a Mach-Zehnder interferometer and drive them simultaneously with multiple and unique acoustic frequencies. As a result, orthogonal light stripes are obtained that interfere with the sample plane, forming a two-dimensional array of flickering spots - each with its modulation frequency. The light from the sample is collected with a single photodiode that, after spectrum analysis, allows for image reconstruction at speeds only limited by the AOD's bandwidth and laser power. We describe the working principle of our approach, characterize its imaging performance as a function of the number of pixels - up to 400 × 400 - and characterize dynamic events at 5000 fps.

Influence of Image Discretization and Patch Size on Microbubble Localization Precision.

For ultrasound localization microscopy, the localization of microbubbles (MBs) is an essential part to obtain super-resolved maps of the vasculature. This paper analyzes the impact of image discretization and patch size on the precision of different MB localization methods to reconcile different observations from previous studies, provide an estimate of feasible localization precision, and derive guidelines for an optimal parameter selection. For this purpose, images of MBs were simulated with Gaussian point-spread functions (PSF) of varying width parameter σ at randomly generated subpixel positions, and Rician distributed noise was added. Four localization methods were tested on patches of different sizes (number of pixels N × N): Gaussian fit, radial symmetry method, calculation of center of mass, and peak detection. Additionally, the Cramér-Rao lower bound (CRLB) for the given estimation problem was calculated. Our results show that the localization precision is strongly influenced by the ratio of the PSF width parameter σ to the pixel size Δ, as well as the patch size N. The best parameter combination depends on the localization method. Generally, very small σ/Δ ratios as well as large σ/Δ ratios in combination with small N lead to performance degradation. The Gaussian fit as representative of a model-based fit comes close to the CRLB and always performs best for the σ/Δ ratios given by image discretization if N is adapted to the PSF. To achieve good results with the Gaussian fit and the radial symmetry method, a good rule of thumb is to set the pixel sizes Δ ≤ σ/0.6 and the patch sizes N ≥ 2σ/Δ + 3.

EVALUATION OF THE OPTIC NERVE IN MIGRAINE PATIENTS BY MAGNETIC RESONANCE IMAGING HISTOGRAM ANALYSIS

Background: Migraine is a chronic, paroxysmal neurovascular primary headache disease seen in approximately 15% of the world. In many studies on migraine, changes in the gray and white matter at the microvascular level have been detected. 90% of patients with aura have visual auras. Histogram analysis is an artificial intelligence application that can detect microscopic pathology in tissues before visual pathology. Objective: Histogram analysis was utilized in our investigation to ascertain whether or not the optic nerve, which is ordinarily visible on magnetic resonance images, was impacted. Methods: 120 adult migraine patients and 120 control group were involved. MRI was performed on a Philips 1.5 Tesla Ingenia device. For histogram analysis, optic nerve region of interest (ROI) measurement was performed in the coronal planes from the edges of the intraorbital segment, whose borders can be clearly distinguished, by using the manual drawing tool for both eyes. The data were analyzed through the MatLab program. Results: The average age of individuals diagnosed with migraine was 37.04 ± 9.57 years, whereas the control group had an average age of 37.8 ± 10.8 years. Between the two groups, 27 distinct parameters computed via histogram analysis were compared. In the control and migraine groups, no statistically significant distinction was observed between the right and left eyes. In contrast to the migraine and control groups, the Mean Local Standard Deviation of the Matrix was found to be The histogram's 5th, 10th, 25th, and 75th percentiles, the average squared, the root sum of squared, the size percent M, the most frequently used value, the median, and the average are all quantified using the root sum of squares. The number of pixels is also indicated. It was determined that ROI values were statistically significant (p0.05). Conclusion: In migraine patients, alterations in the optic nerve, which are typically discernible on conventional magnetic resonance images, may be identified via histogram analysis.

Comparative Protein Profiling of Blood and Milk of Early Pregnant Buffaloes Using SDS-PAGE

Conventional methods for detecting pregnancy in buffalo are inefficient before 30-days after artificial insemination. A proteomic approach for high-resolution analysis of blood plasma and milk proteins by SDS polyacrylamide gel electrophoresis (PAGE) is of great importance. Blood and milk samples were collected on days 0 (before insemination) and 20 and 25 after insemination from retrospective pregnant (n = 6) and non-pregnant (n = 6) animals. An equal volume of samples was loaded in (12%-4%) SDS gel to resolve major plasma and milk proteins. Plasma proteins showed three distinct zones – high, medium, and low molecular weight (MW). In the high MW zone (135 - 250 kDa), ~208 kDa and ~190 kDa proteins were visible. In the moderate MW zone (60 -75 kDa), 5-6 proteins like albumin (~70 kDa), 75 kDa, and 63 kDa consisting of pregnancy-associated glycoproteins, ~73 kDa being dominant. In the low MW zone (25 – 35 kDa), a clear band of 27 kDa was visible. Proteins in milk supernatants showed four distinct zones of proteins – very high, high, medium, and low molecular weight (MW) zones. In the very high MW zone (135 - 250 kDa), ~208 kDa and ~190 kDa proteins were visible. In the high MW zone (60 -75 kDa), proteins like lactoferrin (~78.2 kDa), 73 kDa, albumen (~66 kDa), and a heavy chain of immunoglobulin (IgG) (~54 kDa) were detected. In the medium MW zone (25 – 35 kDa), clear bands of milk caseins, namely thick alpha S2 casein (29 kDa), and other caseins. In the low MW zone (11 – 20 kDa), two thick bands of 18 kDa and 12 kDa were visible. The mean gray values (sum of the gray values of all the pixels in the peak divided by the number of pixels) of the 73 kDa protein peak were analyzed in two-way ANOVA using GraphPad (ver 8.0). Repeated measures of ANOVA were selected where pregnancy status (pregnant and non-pregnant) defined column factor and time (0-, 20- and 25-day) defined row factor. The expression of 73 kDa protein in blood plasma was higher (P = 0.04) in pregnant women than in non-pregnant buffaloes at 20- and 25-days after insemination. However, in milk, there was no difference in the expression of the 73 kDa protein between pregnant and non-pregnant buffaloes across time. Variation in animals within the groups was significantly high. The differential expression of ~73 kDa protein in plasma and precise identification of the protein can be a diagnostic marker for early pregnancy detection in buffaloes.

Improved dark channel priori single image defogging technique using image segmentation and joint filtering.

Foggy images affect image analysis and measurement because of their low definition and blurred details. Despite numerous studies on haze in natural images in hazy environments, the recovery effect is not ideal for processing hazy images in sky areas. A dark channel priori technique for processing haze images with sky areas where atmospheric light values are misestimated and halo artefacts are produced, as well as an improved dark channel priori single-image defogging technique based on image segmentation and joint filtering, are proposed. First, an estimation method of the atmospheric illumination value using image segmentation is proposed to obtain the atmospheric illumination value. The probability density distribution function of the haze-grey image was constructed during image segmentation. The probability density distribution function of the grey image value, the K-means clustering technique, and the method for estimating atmospheric illumination values are combined to improve image segmentation techniques and achieve the segmentation of sky and non-sky areas in hazy images. Based on the segmentation threshold, the number of pixels in the sky and non-sky areas, as well as the normalisation results, were counted to calculate the atmospheric illumination values. Second, to address the halo artefact phenomenon, a method for optimising the image transmittance map using joint filtering is proposed. The image transmittance map was optimised by combining fast-guided filtering and weighted least-squares filtering to retain the edge information and smooth the gradient change of the internal region. Finally, gamma correction and automatic level optimisation are used to improve the brightness and contrast of the defogged images. The experimental results show that the proposed technique can effectively achieve sky segmentation. Compared to the traditional dark-channel prior technique, the proposed technique suppress halo artefacts and improve image detail recovery. Compared to other techniques, the proposed technique exhibited excellent performance in subjective and objective evaluations.

An Efficient VLSI Architecture for Bi-Cubic Interpolation using Carry Skip Adder

In any signal processing system, signal distortion can occur. A compensation scheme would adjust the captured signal to correct for any deviations and ensure accurate signal transmission and preservation. A compensation scheme is used to correct such deviations for mostly linear is used to enhance the quality of reconstructed signal suitable for high-quality super-resolution techniques. This technique is applicable for multidimensional signal processing such as image processing. Except for the resolution of the acquired images, the extracting of colour information is greatly affected by the number of pixels. This scheme enhances the correlation of interpolated pixels and their immediate neighbouring pixels. This research aims to develop a super-resolution demosaicking (SRD) technique to reconstruct high-resolution, full-colour images from samples of the same previously seen (in-painted) image without a sophisticated, time-consuming training process. Hardware-oriented colour demosaicking methods, which give importance to green colour pixels, have thus enhanced the quality of the restored image. A Boundary Mirror and Detector mechanism also considered here for improving the pixel quality up to the boundaries. Three interpolators were economized by using a method called hardware sharing. The present Bi-Cubic interpolation technique has been designed as part of our overall effort to enhance and advance the SRD approach toward better performance and broader acceptance. This had been implemented by VLSI architecture and VHDL, whereas MATLAB had executed the image conversion and de-conversion methods. Comparison in the evaluation of results was done based on PSNR and SSIM values, followed by the analysis of the output results after the synthesis process using a Xilinx Vivado ZYNQ-7-ZC702 FPGA.

Intelligent Brain Tumor Detection and Classification Using Image Processing

The abnormal growths of cells in the brain are called tumours and cancer is a term used to represent malignant tumours. Usually, CT or MRI scans are used for the detection of cancer regions in the brain. Positron Emission Tomography, Cerebral Arteriogram, Lumbar Puncture, Molecular testing are also used for brain tumour detection. In this study, MRI scan images are taken to analyse the disease condition. Objective this research works is i) identify the abnormal image ii) segment tumour region. Density of the tumour can be estimated from the segmented mask and it will help in therapy. Deep learning technique is employed to detect abnormality from MRI images. Multi-level thresholding is applied to segment the tumour region. Number of malignant pixels gives the density of the affected region. The objective of the research is to find the tumour portion of the abnormal MRI brain image using automatic segmentation. The automatic segmentation is accomplished using wavelet transform to extract various features. Then the abnormal images are processed by conventional K means clustering and Fuzzy C means algorithm. But the conventional methods take long iteration and long time to converge. The hybrid clustering method overcomes the long iteration and time.

Click-pixel Cognition Fusion Network with Balanced Cut for Interactive Image Segmentation.

Interactive image segmentation (IIS) has been widely used in various fields, such as medicine, industry, etc. However, some core issues, such as pixel imbalance, remain unresolved so far. Different from existing methods based on pre-processing or post-processing, we analyze the cause of pixel imbalance in depth from the two perspectives of pixel number and pixel difficulty. Based on this, a novel and unified Click-pixel Cognition Fusion network with Balanced Cut (CCF-BC) is proposed in this paper. On the one hand, the Click-pixel Cognition Fusion (CCF) module, inspired by the human cognition mechanism, is designed to increase the number of click-related pixels (namely, positive pixels) being correctly segmented, where the click and visual information are fully fused by using a progressive three-tier interaction strategy. On the other hand, a general loss, Balanced Normalized Focal Loss (BNFL), is proposed. Its core is to use a group of control coefficients related to sample gradients and forces the network to pay more attention to positive and hard-to-segment pixels during training. As a result, BNFL always tends to obtain a balanced cut of positive and negative samples in the decision space. Theoretical analysis shows that the commonly used Focal and BCE losses can be regarded as special cases of BNFL. Experiment results of five well-recognized datasets have shown the superiority of the proposed CCF-BC method compared to other state-of-the-art methods. Code will be publicly available at https://github.com/lab206/CCF-BC.

ПРИМЕНЕНИЕ МЕТОДА ЦВЕТОВОЙ СЕГМЕНТАЦИИ В ЗАДАЧЕ ОПРЕДЕЛЕНИЯ ОДНОРОДНОСТИ КОРМОСМЕСИ

The prepared compound feed’s quality by the resulting mixture uniformity is assessed. The known methods for this indicator determining by the control component or mixture components’ electrophysical properties involve sampling analysis, that is resource expensive and ineffective. The machine vision using for mixing quality in real time determining is more efficient way. However, the feed mixture is blurred bulk media by its characteristics, that makes it difficult photo and video recording to mixture’s composition study carrying out. This study peculiarity is that the mixture’s mixing quality by machine vision methods using to analyze of images obtained blurred bulk media using by an endoscope with an annular illuminator is determined. The endoscope is mounted in a cylindrical protective box with a transparent end glass (the optical axis of the endoscope and the cylinder’s axis are coincided). To analyze each image, a binary mask highlighting only the image’s informative part is constructed it means that image’s part from which the highlights from the annular illuminator and the protective box’s glass outside area are excluded. Next, these colors’ average values of the template image corresponding to the recognized mixture components are calculated, then the number of pixels closest to Euclidean metric to selected color clusters corresponding to the number of mixture’s components is determined. The color’s numerical characteristic by the RGB channels or Lab color representations is set. The paper shows that in the case of a multicomponent mixture working the Lab space gives better results in compared to the RGB space. The estimation of the color segmentation method’s error determining the number of pixels corresponding to a particular color’s cluster is given. It is shown that the morphological operation of image’s connected areas removing reduces the number of "false" pixels in the color cluster. In the future, the obtained results in the feed components uniformity distribution’s program determining will be included.

The Evaluation of Entropy-based Algorithm towards the Production of Closed-Loop Edge

This research concerns the common problem of edge detection that produces a disjointed and incomplete edge, leading to the misdetection of visual objects. The entropy-based algorithm can potentially solve this problem by classifying the pixel belonging to which objects in an image. Hence, the paper aims to evaluate the performance of entropy-based algorithm to produce the closed-loop edge representing the formation of object boundary. The research utilizes the concept of Entropy to sense the uncertainty of pixel membership to the existing objects to classify pixels as the edge or object. Six entropy-based algorithms are evaluated, i.e., the optimum Entropy based on Shannon formula, the optimum of relative-entropy based on Kullback-Leibler divergence, the maximum of optimum entropy neighbor, the minimum of optimum relative-entropy neighbor, the thinning of optimum entropy neighbor, and the thinning of optimum relative-entropy neighbor. The experiment is held to compare the developed algorithms against Canny as a benchmark by employing five performance parameters, i.e., the average number of detected objects, the average number of detected edge pixels, the average size of detected objects, the ratio of the number of edge pixel per object, and the average of ten biggest sizes. The experiment shows that the entropy-based algorithms significantly improve the production of closed-loop edges, and the optimum of relative-entropy neighbor based on Kullback-Leibler divergence becomes the most desired approach among others due to the production of more considerable closed-loop edge in the average. This finding suggests that the entropy-based algorithm is the best choice for edge-based segmentation. The effectiveness of Entropy in the segmentation task is addressed for further research.

The Evaluation of Entropy-based Algorithm towards the Production of Closed-Loop Edge

This research concerns the common problem of edge detection that produces a disjointed and incomplete edge, leading to the misdetection of visual objects. The entropy-based algorithm can potentially solve this problem by classifying the pixel belonging to which objects in an image. Hence, the paper aims to evaluate the performance of entropy-based algorithm to produce the closed-loop edge representing the formation of object boundary. The research utilizes the concept of Entropy to sense the uncertainty of pixel membership to the existing objects to classify pixels as the edge or object. Six entropy-based algorithms are evaluated, i.e., the optimum Entropy based on Shannon formula, the optimum of relative-entropy based on Kullback-Leibler divergence, the maximum of optimum entropy neighbor, the minimum of optimum relative-entropy neighbor, the thinning of optimum entropy neighbor, and the thinning of optimum relative-entropy neighbor. The experiment is held to compare the developed algorithms against Canny as a benchmark by employing five performance parameters, i.e., the average number of detected objects, the average number of detected edge pixels, the average size of detected objects, the ratio of the number of edge pixel per object, and the average of ten biggest sizes. The experiment shows that the entropy-based algorithms significantly improve the production of closed-loop edges, and the optimum of relative-entropy neighbor based on Kullback-Leibler divergence becomes the most desired approach among others due to the production of more considerable closed-loop edge in the average. This finding suggests that the entropy-based algorithm is the best choice for edge-based segmentation. The effectiveness of Entropy in the segmentation task is addressed for further research.