Color-based Segmentation Method Research Articles

Image Processing Technique on Identification of Leaf Types and Detection of Diseases Portion on Tomato Leaves

This study presents novel techniques for leaf type identification, employing color histogram and edge histogram approaches. A color histogram serves as a model to represent color through intensity values, while each image is associated with a descriptive caption. Signatures, encompassing shape, color, and texture, provide a basis for comparing images. The edge histogram, delineating the distribution of five edge types in localized sub-images, further enhances the identification process. To address disease detection on tomato leaves, a color-based segmentation method utilizing the k-means clustering technique is proposed. This iterative approach partitions images into k clusters, facilitating the identification of diseases affecting tomato plants. Beyond environmental factors like rain and temperature, crop diseases emerge as primary influencers on production quality and crop yield. Early detection of diseases is crucial for effective control and mitigation. Leveraging technological advancements, the paper emphasizes the potential of using images of diseased leaves for accurate disease identification. This involves feature extraction from images, which can be subsequently employed in classification algorithms or content-based image retrieval systems.

Testing a Vision-Based Autonomous Drone Navigation Model in a Forest Environment

Drones play a pivotal role in various industries of Industry 4.0. For achieving the application of drones in a dynamic environment, finding a clear path for their autonomous flight requires more research. This paper addresses the problem of finding a navigation path for an autonomous drone based on visual scene information. A deep learning-based object detection approach can localize obstacles detected in a scene. Considering this approach, we propose a solution framework that includes masking with a color-based segmentation method to identify an empty area where the drone can fly. The scene is described using segmented regions and localization points. The proposed approach can be used to remotely guide drones in dynamic environments that have poor coverage from global positioning systems. The simulation results show that the proposed framework with object detection and the proposed masking technique support drone navigation in a dynamic environment based only on the visual input from the front field of view.

Detection of chalcopyrite in digital images of copper ore sample sections

Copper ores are mineral formations of natural origin, the concentration of copper or copper compounds in which is sufficient for their commercial mining. Because of this, copper ores are on the list of minerals that are of strategic importance for the sustainable development of the economy and defense capability of Ukraine. In addition, due to the high thermal conductivity, copper and its alloys are widely used for the manufacture of various types of heat exchangers and radiators. The content of copper in ores usually ranges from 1 to 5%; ores containing less than 0.5% copper are unprofitable for processing at the current level of technology. Mining of copper is of strategic importance, as copper is the main material for the production of cable products and other conductive parts in the electrical industry. The efficiency of technological processes in the mining and industry during the production of copper depends on the quality of ores in relation to the content of minerals that contain copper. Copper ore can be considered a heterogeneous material consisting of two homogeneous phases, one phase is host rock, and the other is copper mineral. Modern image processing techniques have allowed as to automate the identification of minerals in ore samples. Automatic recognition and quantification of minerals using X-Ray tomography, scanning electronic microscope, light microscopy, is one of the most important problems in ore processing systems, as the amount of the minerals in the ore must be determined for further processing. A method of segmentation of colour images of sections of test samples of copper ore is proposed to estimate the percentage content of its components in the section of the tested samples. It is based on the use of chalcopyrite colour features in the HSV model. This colour-based segmentation method is proposed to exploit the average value and distribution of HSV colour components of chalcopyrite in an copper ore image. Segmentation parameters are configured. The experimental results of the segmentation of colour images of copper ore slices by the proposed method are analyzed. The effectiveness of the method is checked using synthesized test images. The method provides an absolute error less than 2.5%.

A Novel Color-Based Segmentation Method for the Objective Measurement of Human Masticatory Performance

The aims of this study were to propose an automatic color-based segmentation method to separate mixed and unmixed colors of images that were derived from the application of the two-color chewing-gum mixing test and to determine the validity of this method in the assessment of masticatory performance (MP). Fifty young adults (mean age: 24.3 ± 2.7 years) were enrolled in the study. Each participant chewed a double-colored chewing gum for 5, 10, 20, 30, and 50 masticatory cycles. Boluses were collected and flattened. Both sides of each bolus were photographed, and images were processed using a novel k-means clustering method. The specimens corresponding to 20 masticatory cycles were re-analyzed by the same investigator in order to evaluate the intra-rater reliability and by a second investigator to assess the inter-rater reliability. To assess the test–retest reliability, 25% of the participants performed a second test with 20 chewing cycles. Each bolus was subjectively scored as either poorly, moderately, or highly mixed by an investigator to assess the construct validity. The percentage of mixed colors in the samples increased with an increase in the number of strokes. Significative differences were detected when varying from 5 to 10 strokes, from 10 to 20 strokes, and from 30 to 50 strokes (p < 0.05). The Pearson correlation coefficient explained these relations (r = 0.78, p < 0.05). The interclass correlation coefficient (ICC) showed a good correlation concerning both the intra- and inter-rater reliability (r = 0.85 and r = 0.77, respectively) and an excellent test–retest correlation (r = 0.93). The subjective assessment was coherent with the digital one. The proposed digital method was proved to be able to automatically quantify the percentage of the mixed color area by providing quantitative data with minimal human interaction.

Feature extraction from high resolution satellite images using K-means and colour threshold approach

In recent year developments in satellite sensors tend to the availability of high spatial and spectral resolution images. The motivation of this research paper is to obtain maximum benefits of different bands from high resolution satellite images in order to put into practice an image processing algorithm solution for extraction and classification of land cover and manmade objects can be used by non-professionals. In this research paper a novel approach for image classification is presented by applying k-means algorithm and colour threshold approach onto high resolution World View 2 (WV2) image. K-means algorithm is applied on a reflectance image to extract land cover classes and manmade objects based on a colour-based segmentation method. The proposed technique is applied through MATLAB environment. The user is asked to select few points of the desired classes and the algorithm do the rest and produce vector layers of the selected classes. The experimental results prove the effectiveness of our framework to enhance the quality of classification in aspects of computational time and precision. The preliminary results are considered promising.

Erythrocyte Fraction Within Retrieved Thrombi Contributes to Thrombolytic Response in Acute Ischemic Stroke.

Recent advent of endovascular thrombectomy (EVT) enables us to provide a new perspective on the use of tPA (tissue-type plasminogen activator) through histological analysis of retrieved thrombus. We investigated the responsiveness of intravenous thrombolysis (IVT) according to the thrombus composition in EVT-attempted patients with acute ischemic stroke. We reviewed 92 consecutive patients with anterior circulation stroke who received combined IVT and EVT for 2 years. IVT responsiveness is defined as any decrease in the clot burden from baseline computed tomographic angiography to digital subtraction angiography during EVT. We histologically analyzed the relative fractions of red blood cells (RBCs), congregated fibrin and platelets, and white blood cells in the retrieved thrombi using semiautomated color-based segmentation method. Clinical characteristics according to the RBC fraction were investigated, and associated factors with IVT responsiveness were explored. Fifty-two patients with histological analyses were stratified into lowest, middle, and highest tertiles of RBC fraction. Toward higher RBC fraction, there was more common susceptibility vessel signs on magnetic resonance imaging (50.0% versus 66.7% versus 91.7%; P=0.022) and prevalent IVT responsiveness (25.0% versus 41.7% versus 75.0%; P=0.010). IVT-responsive group (n=23) had higher RBC fraction (45.7±15.5% versus 35.9±12.2%; P=0.010), lower fibrin and platelet (50.4±14.0% versus 58.5±11.1%; P=0.027), and lower white blood cells fraction (3.9±2.1% versus 5.5±3.0%; P=0.027) than IVT-unresponsive group (n=29). After adjusting for potential variables, RBC fraction (odds ratio, 1.05; 95% confidence interval, 1.01-1.10) remained only independent determinant of IVT responsiveness. In EVT-attempted patients with acute ischemic stroke, IVT responsiveness would be closely associated with RBC fraction.

A study on human gait dynamics: modeling and simulations on OpenSim platform

The analysis of human gait dynamics allows an individual to obtain interesting biometric features through which gait disturbances can be observed from normal and abnormal gait patterns. The musculo-skeletal modeling of human movement helps to study the gait dynamics with expected simulations. It encourages the clinicians to identify the pathological gait for proper treatment. We have created a vision-based human locomotion laboratory to capture the healthy and non-healthy gait patterns. A low cost black uniform is taken to capture different person’s gait data. This dress is made up with a piece of multiple color ribbon to identify different locations of body joints. It provides an added advantage during separation of different joint locations through color-based segmentation method. The demographic information of each subject helps to understand their gait dynamics. It instigates us to build a musculo-skeletal gait model on OpenSim simulation framework. The Regression Analysis (RA) technique is applied on crouch and healthy gait feature to measure the generalization ability and to uncover the gait profiles which are estimated by different error metrics such as, Root Mean Square Error (RMSE), Standard Deviation of Error (SDE), and Mean Error (ME). A computational method based on Normalized Auto Correlation (NAC) is computed to measure the gait disturbances in training subjects. The performance analysis of regression model on motion captured data has been validated with subject specific musculo-skeletal gait model on OpenSim platform.

How to Quantify Penile Corpus Cavernosum Structures with Histomorphometry: Comparison of Two Methods.

The use of morphometrical tools in biomedical research permits the accurate comparison of specimens subjected to different conditions, and the surface density of structures is commonly used for this purpose. The traditional point-counting method is reliable but time-consuming, with computer-aided methods being proposed as an alternative. The aim of this study was to compare the surface density data of penile corpus cavernosum trabecular smooth muscle in different groups of rats, measured by two observers using the point-counting or color-based segmentation method. Ten normotensive and 10 hypertensive male rats were used in this study. Rat penises were processed to obtain smooth muscle immunostained histological slices and photomicrographs captured for analysis. The smooth muscle surface density was measured in both groups by two different observers by the point-counting method and by the color-based segmentation method. Hypertensive rats showed an increase in smooth muscle surface density by the two methods, and no difference was found between the results of the two observers. However, surface density values were higher by the point-counting method. The use of either method did not influence the final interpretation of the results, and both proved to have adequate reproducibility. However, as differences were found between the two methods, results obtained by either method should not be compared.

Epithelium percentage estimation facilitates epithelial quantitative protein measurement in tissue specimens

BackgroundThe rapid advancement of high-throughput tools for quantitative measurement of proteins has demonstrated the potential for the identification of proteins associated with cancer. However, the quantitative results on cancer tissue specimens are usually confounded by tissue heterogeneity, e.g. regions with cancer usually have significantly higher epithelium content yet lower stromal content.ObjectiveIt is therefore necessary to develop a tool to facilitate the interpretation of the results of protein measurements in tissue specimens.MethodsEpithelial cell adhesion molecule (EpCAM) and cathepsin L (CTSL) are two epithelial proteins whose expressions in normal and tumorous prostate tissues were confirmed by measuring staining intensity with immunohistochemical staining (IHC). The expressions of these proteins were measured by ELISA in protein extracts from OCT embedded frozen prostate tissues. To eliminate the influence of tissue heterogeneity on epithelial protein quantification measured by ELISA, a color-based segmentation method was developed in-house for estimation of epithelium content using H&E histology slides from the same prostate tissues and the estimated epithelium percentage was used to normalize the ELISA results. The epithelium contents of the same slides were also estimated by a pathologist and used to normalize the ELISA results. The computer based results were compared with the pathologist’s reading.ResultsWe found that both EpCAM and CTSL levels, measured by ELISA assays itself, were greatly affected by epithelium content in the tissue specimens. Without adjusting for epithelium percentage, both EpCAM and CTSL levels appeared significantly higher in tumor tissues than normal tissues with a p value less than 0.001. However, after normalization by the epithelium percentage, ELISA measurements of both EpCAM and CTSL were in agreement with IHC staining results, showing a significant increase only in EpCAM with no difference in CTSL expression in cancer tissues. These results were obtained with normalization by both the computer estimated and pathologist estimated epithelium percentage.ConclusionsOur results show that estimation of tissue epithelium percentage using our color-based segmentation method correlates well with pathologists' estimation of tissue epithelium percentages. The epithelium contents estimated by color-based segmentation may be useful in immuno-based analysis or clinical proteomic analysis of tumor proteins. The codes used for epithelium estimation as well as the micrographs with estimated epithelium content are available online.

Image Segmentation and Feature Extraction Using Sift-Sad Algorithm for Disparity Map Generation

In this paper, a stereo matching algorithm based on image segments and disparity measurement using stereo images is presented. We propose the hybrid segmentation algorithm that is based on a combination of the Belief Propagation and Mean Shift algorithms with aim to refine the final disparity map by using a stereo pair of images. Firstly, a color based segmentation method is applied for segmenting the left image of the input stereo pair (reference image) into regions. The aim of the segmentation is to simplify representation of the image into the form that is easier to analyze and is able to locate objects in images. Secondly, results of the segmentation are used as an input of the SIFTSAD matching method to determine the disparity estimate of each image pixel. This matching algorithm is proposed by combining Scale Invariant Feature Transform (SIFT) with the Sum of Absolute Difference (SAD). Finally, the comparisons between the three robust feature detection methods SIFT, Affine SIFT (ASIFT) and Speeded Up Robust Features (SURF) are presented. The obtained experimental results demonstrate that the proposed method has a positive effect on overall estimation of disparity map and outperforms other examined methods.