Blob Analysis Research Articles

Research on online visual recognition of coal gangue based on BLOB analysis and support vector machine

Research on online visual recognition of coal gangue based on BLOB analysis and support vector machine

Understanding Kirkendall effect in Ni(W) diffusion-induced recrystallization region

This study elucidates the formation of distinct phases, such as intermetallic compounds (IMCs), and the evolution of Kirkendall voids. We not only observed the emergence of polycrystalline IMC layers and distinct diffusion-induced recrystallization (DIR) regions using various experimental methods but also quantified the elusive Kirkendall vacancy fraction by an extended numerical model. Experiments revealed the formation of polycrystalline layers with Ni4W and a distinct DIR region. Irregularly shaped Kirkendall voids were found in the DIR region in the Ni matrix. The observed DIR region thickness exceeding W penetration depth indicated rapid alloying. Experimental data and insights from the Ni-W phase diagram showed exclusive Ni4W formation. EBSD analysis revealed smaller Ni4W grains and larger, irregularly shaped DIR region grains. Quantification of Kirkendall voids through Blob analysis linked numerical results to experiments. Noteworthy findings include the fraction of Kirkendall vacancy and the evolution of the possible vacancy region calculations, highlighting their dependence on the W diffusivity and grain boundary diffusion. Temporal evolution of the vacancy region showed the absence of vacancies near Ni4W/DIR, with the leading edge at DIR/Ni, providing insights into annealing-induced vacancy distribution. This integrated approach enhances understanding of thermodynamics and kinetics in the Ni-W diffusion couple, advancing knowledge crucial for high-temperature materials science applications.

Research on the examination technology of connector pin skewing according to Blob analysis *

Pin skew detection is an important means to ensure the reliable operation of connectors. To address the issues of low accuracy and limited applicability in existing research, this paper proposes a connector pin skew detection method based on Blob analysis. Firstly, the image is segmented by incorporating the dimensional features of the tested connector to retain the effective information region in the image, reducing the computational workload for subsequent image processing. The image is preprocessed using an improved median filtering algorithm to effectively mitigate the interference of noise on the detection process. Secondly, a locally adaptive approach is employed to dynamically adjust the threshold, and morphological processing is applied to the pin image to enhance the pin speckle features. Subsequently, Blob analysis is utilized to analyze the connector pin speckles, obtaining data on the pin skew. Different evaluation criteria for pin skew data of various connectors are established to achieve quantitative assessment. Finally, experiments are conducted for pin skew detection of single-hole rectangular, double-hole rectangular, and single-hole circular connectors. The experimental results demonstrate that the proposed connector pin skew detection method can effectively detect various types of pin skew in connectors, with a detection accuracy better than 0.05 mm and a repeatability better than 0.03 mm. This method is suitable for automatic detection scenarios of connector pin skew.

GRD-Net: Generative-Reconstructive-Discriminative Anomaly Detection with Region of Interest Attention Module

Anomaly detection is nowadays increasingly used in industrial applications and processes. One of the main fields of the appliance is the visual inspection for surface anomaly detection, which aims to spot regions that deviate from regularity and consequently identify abnormal products. Defect localization is a key task that is usually achieved using a basic comparison between generated image and the original one, implementing some blob analysis or image-editing algorithms in the postprocessing step, which is very biased towards the source dataset, and they are unable to generalize. Furthermore, in industrial applications, the totality of the image is not always interesting but could be one or some regions of interest (ROIs), where only in those areas there are relevant anomalies to be spotted. For these reasons, we propose a new architecture composed by two blocks. The first block is a generative adversarial network (GAN), based on a residual autoencoder (ResAE), to perform reconstruction and denoising processes, while the second block produces image segmentation, spotting defects. This method learns from a dataset composed of good products and generated synthetic defects. The discriminative network is trained using a ROI for each image contained in the training dataset. The network will learn in which area anomalies are relevant. This approach guarantees the reduction of using preprocessing algorithms, formerly developed with blob analysis and image-editing procedures. To test our model, we used challenging MVTec anomaly detection datasets and an industrial large dataset of pharmaceutical BFS strips of vials. This set constitutes a more realistic use case of the aforementioned network.

A Simplified Experimental Method to Estimate the Transport of Non-Buoyant Plastic Particles Due to Waves by 2D Image Processing

Plastic is one of the most widely used materials. It is expected that there will be about 12 billion tons of debris dispersed into the environment by 2050. This, combined with marine littering, pollution and climate change will threaten our ability to sustainably use oceans, seas and coasts. In this regard, it is important to understand transport mechanisms and predict hotspots of plastic pollution. Physical models are a valid support to achieve such a goal. In this regard, it is necessary to adopt specific techniques which allow the movement of plastic to be monitored. Considering the relatively high size of plastic particles that could be used during experimental campaigns, new approaches based on image processing could be implemented. This study presents a simplified method to estimate the transport of particle debris under sea waves by 2D image processing, avoiding any disturbance of the dynamic field. The proposed method, based on the blob analysis, was implemented and calibrated within the wave flume of the Hydraulics Laboratory at the University of Messina. After a calibration process, the method was used to study the behaviour of the plastic debris under several hydrodynamic conditions. The results obtained in terms of displacement and velocity of the analysed sample showed a great reliability of the used methodology.

Precision landing comparison between smartphone video guidance sensor and IRLock by hardware-in-the-loop emulation

The Smartphone Video Guidance Sensor (SVGS) is an emerging technology developed by NASA Marshall Space Flight Center that uses a vision-based approach to accurately estimate the six-state position and orientation vectors of an illuminated target of known dimensions with respect to a coordinate frame fixed to the camera. SVGS is a software-based sensor that can be deployed using a host platform’s resources (CPU and camera) for proximity operations and formation flight of drones or spacecraft. The SVGS output is calculated based on photogrammetric analysis of the light blobs in each image; its accuracy in linear and angular motion at different velocities has previously been successfully demonstrated [9]. SVGS has several potential applications in guidance, navigation, motion control, and proximity operations as a reduced-cost, compact, reliable option to competing technologies such as LiDAR or infrared sensing. One of the applications envisioned by NASA for SVGS is planetary/lunar autonomous landing. This paper aims to compare the SVGS performance in autonomous landing with existing technologies: a combination of infrared beacon technology (IRLock) and LiDAR. The comparison is based on a hardware-in-the-loop emulation of a precision landing experiment using a computer-controlled linear motion to emulate the approach motion and the ROS/Gazebo environment to emulate the response of the flight controller to the environment during landing. Results suggest that SVGS performs better than the existing IRLock with LiDAR sensor combination.

Automatic defect detection in fiber-reinforced polymer matrix composites using thermographic vision data

The detection of internal defects, not visible to the naked eye from the outside of materials, using non-destructive testing (NDT) are increasingly requested by industrial processes. This study proposes a novel methodology for acquisition and processing of images from a thermographic camera using computer vision methods to test composite materials made of a polymer matrix reinforced with glass, carbon, and kevlar fibers. The image is acquired while cooling the sample, following a suggested procedure. The processing methodology is divided into three steps, image pre-processing, image processing, and data post-processing. In image preprocessing, filters are applied to improve image quality, and methods are proposed to segment and identify the region of interest. In image processing, a blob analysis method is suggested for defect identification, isolation and characterization. A data analysis method is proposed for the post-processing step to characterize the defects identified in the previous step. Samples with known defects in terms of size, geometry, and location were used to test the developed system. The system showed high performance, achieving 98% accuracy, and suitability for defect detection larger than 0.5 mm in thickness and 600 mm2 in area. The experimental results showed that the algorithm did not detect any false positives, and that the type of reinforcement used in the analyzed samples had no influence on the results. On the other hand, the depth of the delaminations had an influence on the pixel intensity contrast of the defect region, and its instant of maximum contrast. The lesser the depth of the defects detected, the higher the value of their intensity and the shorter the instant of maximum contrast.

An Efficient Vehicle Detection and Shadow Removal Using Gaussian Mixture Models with Blob Analysis for Machine Vision Application

An Efficient Vehicle Detection and Shadow Removal Using Gaussian Mixture Models with Blob Analysis for Machine Vision Application

Application access for chromatic color detection

The process of identifying specific examples of an item inside an image or video is referred to as “object detection,” and it is one of the computer vision techniques. When individuals look at pictures or movies, they are able to recognize and find things that are of interest to them. The primary goal of the field of object recognition is to achieve this level of intelligence via the use of a computer. Image segmentation and blob analysis, which uses simple object properties such as size, shape, or color, and Feature-based object detection, which uses feature extraction, matching, and RANSAC to estimate the location of an object, are two of the many common techniques that are used in conjunction with deep learning and machine learning-based object detection. There are also many other common techniques that are used in conjunction with deep learning and machine learning-based object detection. The most challenging aspects of object detection are the processes of categorizing things and pinpointing their locations. When seen from a different perspective, the same object might seem to be completely different. Detection algorithms have a tough time distinguishing between various things at different scales and perspectives since the sizes and proportions of items almost always alter. The camera input would be used to determine an object’s shape, according to this project’s proposal. Circle, square, and triangle are the three distinct shapes that we are determining an item to be right now. The finished result is shown in MATLAB’s graphical user interface (GUI). The productive identification process is carried out with the assistance of MATLAB. Using the SFTA Algorithm, the application extracts features from the specific image of a video. The neural network algorithm is used to classify the image of video. Image segmentation and image detection are two examples of related computer vision methods that are inextricably intertwined with object recognition. These techniques allow us to better comprehend and analyze the situations shown in movies and still photographs. It has been determined whether or not the suggested system is suitable. Because of this, the testing has shown that the suggested system generates generally accurate indications of real performance at a rate of 99.4%.

Objective biomarkers of depression: A study of Granger causality and wavelet coherence in resting-state fMRI.

The lack of a robust diagnostic biomarker makes understanding depression from a neurobiological standpoint an important goal, especially in the context of brain imaging. In this study, we aim to create novel image-based features for objective diagnosis of depression. Resting-state network time series are used to investigate neurodynamics with the help of wavelet coherence and Granger causality (G-causality). Three new features are introduced: total wavelet coherence, wavelet lead coherence, and wavelet coherence blob analysis. The fourth feature, pair-wise conditional G-causality, is used to establish the causality between resting-state networks. We use the proposed features to classify depression in adult subjects. We obtained an accuracy of 86% in the wavelet lead coherence, 80% in Granger causality, and 86% in wavelet coherence blob analysis. Subjects with depression showed hyperconnectivity between the dorsal attention network and the auditory network as well as between the posterior default mode network and the dorsal attention network. Hypoconnectivity was found between the anterior default mode network and the auditory network as well as the right frontoparietal network and the lateral visual network. An abnormal co-activation pattern was found between cerebellum and the lateral motor network according to the wavelet coherence blob analysis. Based on abnormal functional dynamics between brain networks, we were able to identify subjects with depression with high accuracy. The findings of this study contribute to the understanding of the impaired emotional and attention processing associated with depression, as well as decreased motor activity.

A Novel Method for LCD Module Alignment and Particle Detection in Anisotropic Conductive Film Bonding

In this paper, we propose a misalignment correct method and a particle detection algorithm to improve the accuracy in the quality inspection of the LCD module after the anisotropic conductive film (ACF) bonding. We use only one camera to acquire images of multiple positions in order to establish the transformation from the image space to the world coordinate. Our method can accurately determine the center of rotation of the carrier table and calculate the deviation of position and angle of the tested module. Compared to traditional ways that rely on multiple cameras to align the large-sized product, our method has the advantages of simple structure, low cost, and fast calibration process. The particle detection is performed after positioning all bumps of the bonded module. The gray morphology-based algorithm is developed to detect the extreme point of every particle and refine the particle result through blob analysis. This method reduces the over-checking rate and performs better on the detection precision for dense particles. We verify the effectiveness of our proposed methods in our experiments. The alignment error can be less than 0.05 mm, and the accuracy of the particle detection is 93% while the recall rate is 92.4%.

Small Low-Contrast Target Detection: Data-Driven Spatiotemporal Feature Fusion and Implementation.

Detecting small low-contrast targets in the airspace is an essential and challenging task. This article proposes a simple and effective data-driven support vector machine (SVM)-based spatiotemporal feature fusion detection method for small low-contrast targets. We design a novel pixel-level feature, called a spatiotemporal profile, to depict the discontinuity of each pixel in the spatial and temporal domains The spatiotemporal profile is a local patch of the spatiotemporal feature maps concatenated by the spatial feature maps and temporal feature maps in channelwise, which are generated by the morphological black-hat filter and a ghost-free dark-focusing frame difference methods, respectively. Instead of the handcrafted feature fusion mechanisms in previous works, we use the labeled spatiotemporal profiles to train an SVM classifier to learn the spatiotemporal feature fusion mechanism automatically. To speed up detection for high-resolution videos, the serial SVM classification process on central processing units (CPUs) is reformed as parallel convolution operations on graphics processing unit (GPUs), which exhibits over 1000+ times speedup in our real experiments. Finally, blob analysis is applied to generate final detection results. Elaborate experiments are conducted, and experimental results demonstrate that the proposed method performs better than 12 baseline methods for the small low-contrast target detection. The field tests manifest that the parallel implementation of the proposed method can realize real-time detection at 15.3 FPS for videos at a resolution of 2048×1536 and the maximum detection distance can reach 1 km for drones in sunny weather.

Vehicle Counting Method Based On Gaussian Mixture Models And Blob Analysis

Combination of a Gaussian Mixture Models and Entropy to detect vehicles in motion from any sort of video sequences. The proposed method is composed of four major steps: background subtraction, improved GMM, Blob Analysis, Voting Scheme processing. An implementation of proposed technique has been performed using MATLAB R2017a. Vehicle detection process provides relevant information about traffic patterns, crash occurrences and traffic peak times in roadways. Detection of number of vehicles and traffic monitoring will be helpful by implementing motion and object detection using background subtraction, Monitoring traffic from video footages will provide a greater reality in terms of vehicle count. Gaussian Mixture Model is most advanced model in motion detection due to the reliability that it has shown in the background extraction and foreground segmentation process. Here we are going to find motion of object in video sequences under both static camera arrangement and dynamic background sequences by using background subtraction method and blob analysis approach and analyzing them

COVID-DAI: A novel framework for COVID-19 detection and infection growth estimation using computed tomography images.

The COVID‐19 pandemic is spreading at a fast pace around the world and has a high mortality rate. Since there is no proper treatment of COVID‐19 and its multiple variants, for example, Alpha, Beta, Gamma, and Delta, being more infectious in nature are affecting millions of people, further complicates the detection process, so, victims are at the risk of death. However, timely and accurate diagnosis of this deadly virus can not only save the patients from life loss but can also prevent them from the complex treatment procedures. Accurate segmentation and classification of COVID‐19 is a tedious job due to the extensive variations in its shape and similarity with other diseases like Pneumonia. Furthermore, the existing techniques have hardly focused on the infection growth estimation over time which can assist the doctors to better analyze the condition of COVID‐19‐affected patients. In this work, we tried to overcome the shortcomings of existing studies by proposing a model capable of segmenting, classifying the COVID‐19 from computed tomography images, and predicting its behavior over a certain period. The framework comprises four main steps: (i) data preparation, (ii) segmentation, (iii) infection growth estimation, and (iv) classification. After performing the pre‐processing step, we introduced the DenseNet‐77 based UNET approach. Initially, the DenseNet‐77 is used at the Encoder module of the UNET model to calculate the deep keypoints which are later segmented to show the coronavirus region. Then, the infection growth estimation of COVID‐19 per patient is estimated using the blob analysis. Finally, we employed the DenseNet‐77 framework as an end‐to‐end network to classify the input images into three classes namely healthy, COVID‐19‐affected, and pneumonia images. We evaluated the proposed model over the COVID‐19‐20 and COVIDx CT‐2A datasets for segmentation and classification tasks, respectively. Furthermore, unlike existing techniques, we performed a cross‐dataset evaluation to show the generalization ability of our method. The quantitative and qualitative evaluation confirms that our method is robust to both COVID‐19 segmentation and classification and can accurately predict the infection growth in a certain time frame.Research highlights We present an improved UNET framework with a DenseNet‐77‐based encoder for deep keypoints extraction to enhance the identification and segmentation performance of the coronavirus while reducing the computational complexity as well.We propose a computationally robust approach for COVID‐19 infection segmentation due to fewer model parameters.Robust segmentation of COVID‐19 due to accurate feature computation power of DenseNet‐77.A module is introduced to predict the infection growth of COVID‐19 for a patient to analyze its severity over time.We present such a framework that can effectively classify the samples into several classes, that is, COVID‐19, Pneumonia, and healthy samples.Rigorous experimentation was performed including the cross‐dataset evaluation to prove the efficacy of the presented technique

Segmentasi Citra Bunga Menggunakan Blob Analisis

Content Based Image Retrieval (CBIR) for image segmentation is a concern this year, especially in the development of computer vision. The object discussed in this study is about interest, which uses a dataset from ImageCLEF2017 by taking 8 flower samples. Image of flowers in the dataset is still a lot of noise such as the initial background behind objects such as leaves, tree trunks or others. So we need a method to eliminate the noise, this method for cleaning noise is done by color clusters using the K-means method. By color clustering using K-Means and using color clusters k=2,3,4,and5. After that, a morphological process is carried out in order to obtain a clean area so that it can be compared with the original image and the Blob values formed. Blob analysis is calculated after the process of cleaning the noise is done in order to get the best value in the process of recognition of images with objects of interest. The results of the segmentation process that have been done are the highest MSE and RMSE values are at k-means results with k=4, while for PNSR are at k=2, and for the lowest MSE and RMSE values are at k=5, while the lowest PNSR is at k=4

Tractor-trailer-train braking time sequence detection based on monocular vision

The tractor-trailer-train at the braking process prone to braking instability caused by asynchronous braking between the shafts. With respect to the lack of intelligent detection of Braking Time Sequence (BTS), a non-contact dynamic detection scheme of intelligent vehicle BTS is proposed. Based on the monocular vision principle, the edge markers of tractor-trailer train tires are identified, and the tire slip rate is solved. The noise reduction of the collected image is processed. The marker area is obtained by Blob analysis. This region at the image to be matched is identified by the template matching algorithm based on contour. The camera is calibrated by Zhang’s calibration method. In order to verify the effectiveness of the detection scheme, the real vehicle test was carried out. The test results show that the error of slip rate solution is below 4.2%.

Diagnosis of Diabetic Retinopathy using Blob Analysis

Diagnosis of Diabetic Retinopathy using Blob Analysis

Vision-based real-time vehicle detection and vehicle speed measurement using morphology and binary logical operation

In recent trends, digital information to the industrial integration for the intelligent transportation system (ITS) field is gaining importance for the researcher, academia, and industrial persons. Visual information helps to manage traffic systems in the industrial forum to build smart cities in developed countries. This paper presents vision-based real-time vehicle detection and Vehicle Speed Measurement (VSM) using morphology operation and binary logical process for an unplanned traffic scenario using image processing techniques. Vehicle detection and VSM help to reduce the number of accidents and improve road network efficiency. The bounding box size for vehicle detection is flexible according to vehicles' sizes on the road. We test this system with different colors and dimensions for a selected Region of Interest (ROI). The ROI sets using the two-line approach. Here, we compare the proposed system with the inter-frame difference method and the blob analysis method with recall, precision, and F1 performance parameters.

Computer-Aided Somatic Cells Counting on Three Different Milk Staining Condition

The maximum amount of somatic cell in a millilitre of milk defined by National Standardization Agency is 4x105 cells/ml. The condition of milk will be excluded from a decent quality if the amount of somatic cell is greater than it. Microscopic image of somatic cells counting is a conventional method which utilized by farmer or researcher to count the number of somatic cells from dairy milk sample. The problems of this method are: 1) The counting process is manually conducted which prone to miscalculation, 2) The different colour of staining (purplish, bluish, and greenish) may complicate the calculation. In this research, a computer-based approach is proposed to calculate the number of somatic cells from sample of milk and to eliminate the difficulty of different colour due to staining technique. Image processing knowledge i.e. erosion, dilation, colour conversion, and BLOB Analysis are involved and utilized to achieve the objective. Overall, the correctness of the system in detecting the number of somatic cells is 94%. From this study, a comprehensive system to calculate the number of somatic cells from cow’s milk will be closer to be implemented.

Abandoned Object Detection

In the current fast paced life, security has become an important aspect which cannot be ignored. One of the usually neglected but important division of security in public areas involves the unattended objects. Bombs and other harmful objects that are incendiary cause great harm. And in most of the cases such objects are left unattended for some time before causing harm. As such, unattended objects should be given importance and must be checked to ensure safety. This paper shows a method to identify unattended objects in public areas. This is done by comparing a fixed/initial background frame with frames after a fixed interval. If there is a change in the frames, then that refers to an attended object. No change observed, proves an unattended object which is highlighted by a box. This is done with the help of blob analysis. This paper does this with the help of MATLAB. MATLAB being a software accessible to wide range of people provides an easy process with moderate results. It uses a wide range of predefined functions and toolboxes to do this.