Connected Component Analysis Research Articles

An optimized run-length based algorithm for sparse remote sensing image labeling

Labeling of the connected components is the key operation of the target recognition and segmentation in remote sensing images. The conventional connected-component labeling (CCL) algorithms for ordinary optical images are considered time-consuming in processing the remote sensing images because of the larger size. A dynamic run-length based CCL algorithm (DyRLC) is proposed in this paper for the large size, big granularity sparse remote sensing image, such as space debris images and ship images. In addition, the equivalence matrix method is proposed to help design the pre-processing method to accelerate the equivalence labels resolving. The result shows our algorithm outperforms 22.86% on execution time than the other algorithms in space debris image dataset. The proposed algorithm also can be implemented on the field programming logical array (FPGA) to enable the realization of the real-time processing on-board.

Integrated technique of segmentation and classification methods with connected components analysis for road extraction from orthophoto images

Road networks are one of the main urban features. Therefore, road parts extraction from high-resolution remotely sensed imagery and updated road database are beneficial for many GIS applications. However, owing to the presence of various types of obstacles in the images, such as shadows, cars, and trees, with similar transparency and spectral values as road class, achieving accurate road extraction using different classification and segmentation methods is still difficult. This paper proposes an integrated method combining segmentation and classification methods with connected components analysis to extract road class from orthophoto images. The proposed technique is threefold. First, multiresolution segmentation method was applied to segment images. Then, the main classification methods, namely, decision trees (DT), k-nearest neighbors (KNN), and support vector machines (SVM), were implemented based on spectral, geometric, and textural information to classify the obtained results into two classes: road and non-road. Three main accuracy evaluation measures, such as recall, precision, and F1-score, were evaluated to determine the performance of the proposed method, with respective average values of 87.62%, 89.71%, and 88.61%, respectively, for DT; 86.61%, 88.17%, and 87.30%, respectively, for KNN; and 89.83%, 89.52%, and 89.67%, respectively, for SVM. Finally, connected components labelling was used to extract road component parts, and morphological operation was employed to delete non-road parts and noises and improve the performance. These results were also compared with other prior works, which confirmed that the integrated method is an effective road extraction technique.

A New GMRF Self-supervised Algorithm Applied to SAR Image Classification

The problem of limited tagged training samples and unknown the number of classes is challenging for the classification of remote sensing scenes. This paper presents a new GMRF self-supervised algorithm for SAR image. We add a GOF process in the process of estimating GMM parameters by EM algorithm, which can not only dynamically select the best number of significant classes but also provides an initial feature parameter to calculate the MRF minimum energy. After the iterative region label and region growth cycle, iteration is combined with the Mll context model to obtain the best mark of each region. Since the initial feature parameter selection of the MRF is not random, the operation efficiency is also improved while reducing the number of iteration cycles of the algorithm. The experiment validates that our design not only solves the problem of manual input of the number of classes but also provide the better output result graph in terms of detail maintenance than the expert interpretation of the truth map in the unsupervised image classification process, and we hope that it could support operation and meet the real-time requirements.

18F]F13640, a 5-HT1A Receptor Radiopharmaceutical Sensitive to Brain Serotonin Fluctuations.

IntroductionSerotonin is involved in a variety of physiological functions and brain disorders. In this context, efforts have been made to investigate the in vivo fluctuations of this neurotransmitter using positron emission tomography (PET) imaging paradigms. Since serotonin is a full agonist, it binds preferentially to G-protein coupled receptors. In contrast, antagonist PET ligands additionally interact with uncoupled receptors. This could explain the lack of sensitivity to serotonin fluctuations of current 5-HT1A radiopharmaceuticals which are mainly antagonists and suggests that agonist radiotracers would be more appropriate to measure changes in neurotransmitter release. The present study evaluated the sensitivity to endogenous serotonin release of a recently developed, selective 5-HT1A receptor PET radiopharmaceutical, the agonist [18F]F13640 (a.k.a. befiradol or NLX-112).Materials and MethodsFour cats each underwent three PET scans with [18F]F13640, i.e., a control PET scan of 90 min, a PET scan preceded 30 min before by an intravenous injection 1 mg/kg of d-fenfluramine, a serotonin releaser (blocking challenge), and a PET scan comprising the intravenous injection of 1 mg/kg of d-fenfluramine 30 min after the radiotracer injection (displacement challenge). Data were analyzed with regions of interest and voxel-based approaches. A lp-ntPET model approach was implemented to determine the dynamic of serotonin release during the challenge study.ResultsD-fenfluramine pretreatment elicited a massive inhibition of [18F]F13640 labeling in regions known to express 5-HT1A receptors, e.g., raphe nuclei, hippocampus, thalamus, anterior cingulate cortex, caudate putamen, occipital, frontal and parietal cortices, and gray matter of cerebellum. Administration of d-fenfluramine during PET acquisition indicates changes in occupancy from 10% (thalamus) to 31% (gray matter of cerebellum) even though the dissociation rate of [18F]F13640 over the 90 min acquisition time was modest. The lp-ntPET simulation succeeded in differentiating the control and challenge conditions.ConclusionThe present findings demonstrate that labeling of 5-HT1A receptors with [18F]F13640 is sensitive to serotonin concentration fluctuations in vivo. Although the data underline the need to perform longer PET scan to ensure accurate measure of displacement, they support clinical development of [18F]F13640 as a tool to explore experimental paradigms involving physiological or pathological (neurological or neuropsychiatric pathologies) fluctuations of extracellular serotonin.

Multithreaded two-pass connected components labelling and particle analysis in ImageJ.

Sequential region labelling, also known as connected components labelling, is a standard image segmentation problem that joins contiguous foreground pixels into blobs. Despite its long development history and widespread use across diverse domains such as bone biology, materials science and geology, connected components labelling can still form a bottleneck in image processing pipelines. Here, I describe a multithreaded implementation of classical two-pass sequential region labelling and introduce an efficient collision resolution step, ‘bucket fountain’. Code was validated on test images and against commercial software (Avizo). It was performance tested on images from 2 MB (161 particles) to 6.5 GB (437 508 particles) to determine whether theoretical linear scaling (O(n)) had been achieved, and on 1–40 CPU threads to measure speed improvements due to multithreading. The new implementation achieves linear scaling (b = 0.905–1.052, time ∝ pixelsb; R2 = 0.985–0.996), which improves with increasing thread number up to 8–16 threads, suggesting that it is memory bandwidth limited. This new implementation of sequential region labelling reduces the time required from hours to a few tens of seconds for images of several GB, and is limited only by hardware scale. It is available open source and free of charge in BoneJ.

Oil spill extraction from X-band marine radar images by power fitting of radar echoes

ABSTRACT Oil spills cause great environmental damage and the ability to identify them accurately is important to prompt oil spill response. A power fitting method of radar echo is proposed in this paper to detect oil spills on sampled X-band marine radar images. In this method, first, the marine radar image is pre-processed, including coordinate transform and noise reduction. Then, the difference between the pre-processed radar image and power fitting result is obtained. Finally, oil spills are extracted by using a mean filter, the Otsu method, and connected component analysis. The proposed oil spill detection method was used on radar images sampled after an oil spill accident in the coastal region of Dalian, China, on 21 July 2010. The proposed method can detect oil spills without manual participation, and the extracted oil spills are comparable to visual interpretation.

Multiple Kernel Graph Cut for SAR Image Change Detection

Complementary information between two difference images (DI’s) has great contribution to improve change detection performances. Based on the effectiveness and flexibility of the multiple kernel learning (MKL) in information fusion, we develop a multiple kernel graph cut (MKGC) algorithm for synthetic aperture radar (SAR) image change detection. An energy function containing a weighted summation kernel is proposed for fusing the complementary information between the subtraction image and the ratio image. By iteratively minimizing the energy function, the kernel weights, region parameters and region labels are estimated automatically and optimally. Besides of it avoids modeling, MKGC also has a complete description of the changed areas and the strong noise immunity. Experiments on real GaoFen-3 SAR data set demonstrate the effectiveness of the MKGC algorithm, and illustrate that it is a good candidate for SAR image change detection.

Local Feature Extraction Network for Point Cloud Analysis

Geometric feature extraction of 3D point clouds plays an important role in many 3D computer vision applications such as region labeling, 3D reconstruction, object segmentation, and recognition. However, hand-designed features on point clouds lack semantic information, so cannot meet these requirements. In this paper, we propose local feature extraction network (LFE-Net) which focus on extracting local feature for point clouds analysis. Such geometric features learning from a relation of local points can be used in a variety of shape analysis problems such as classification, part segmentation, and point matching. LFE-Net consists of local geometric relation (LGR) module which aims to learn a high-dimensional local feature to express the relation between points and their neighbors. Benefiting from the additional singular values of local points and hierarchical neural networks, the learned local features are robust to permutation and rigid transformation so that they can be transformed into 3D descriptors. Moreover, we embed prior spatial information of the local points into the sub-features for combining features from multiple levels. LFE-Net achieves state-of-the-art performances on standard benchmarks including ModelNet40, ShapeNetPart.

Using a convolutional neural network for human recognition in a staff dose management software for fluoroscopic interventional procedures.

Staff dose management is a continuing concern in fluoroscopically-guided interventional (FGI) procedures. Being unaware of radiation scatter levels can lead to unnecessarily high stochastic and deterministic risks due to the effects of absorbed dose by staff members. Our group has developed a scattered-radiation display system (SDS) capable of monitoring system parameters in real-time using a controller-area network (CAN) bus interface and displaying a color-coded mapping of the Compton-scatter distribution. This system additionally uses a time-of-flight depth sensing camera to track staff member positional information for dose rate updates. The current work capitalizes on our body tracking methodology to facilitate individualized dose recording via human recognition using 16-bit grayscale depth maps acquired using a Microsoft Kinect V2. Background features are removed from the images using a depth threshold technique and connected component analysis, which results in a body silhouette binary mask. The masks are then fed into a convolutional neural network (CNN) for identification of unique body shape features. The CNN was trained using 144 binary masks for each of four individuals (total of 576 images). Initial results indicate high-fidelity prediction (97.3% testing accuracy) from the CNN irrespective of obstructing objects (face masks and lead aprons). Body tracking is still maintained when protective attire is introduced, albeit with a slight increase in positional data error. Dose reports are then able to be produced which contain cumulative dose to each staff member at the eye lens level, waist level, and collar level. Individualized cumulative dose reporting through the use of a CNN in addition to real-time feedback in the clinic will lead to improved radiation dose management.

Sulfonated cryogel scaffolds for focal delivery in ex-vivo brain tissue cultures

The human brain has unique features that are difficult to study in animal models, including the mechanisms underlying neurodevelopmental and psychiatric disorders. Despite recent advances in human primary brain tissue culture systems, the use of these models to elucidate cellular disease mechanisms remains limited. A major reason for this is the lack of tools available to precisely manipulate a specific area of the tissue in a reproducible manner. Here we report an easy-to-use tool for site-specific manipulation of human brain tissue in culture. We show that line-shaped cryogel scaffolds synthesized with precise microscale dimensions allow the targeted delivery of a reagent to a specific region of human brain tissue in culture. 3-sulfopropyl acrylate (SPA) was incorporated into the cryogel network to yield a negative surface charge for the reversible binding of molecular cargo. The fluorescent dyes BODIPY and DiI were used as model cargos to show that placement of dye loaded scaffolds onto brain tissue in culture resulted in controlled delivery without a burst release, and labelling of specific regions without tissue damage. We further show that cryogels can deliver tetrodotoxin to tissue, inhibiting neuronal function in a reversible manner. The robust nature and precise dimensions of the cryogel resulted in a user-friendly and reproducible tool to manipulate primary human tissue cultures. These easy-to-use cryogels offer an innovate approach for more complex manipulations of ex-vivo tissue.

Chip Appearance Inspection Method for High-Precision SMT Equipment

In order to meet the defect-detection requirements of chips in high-precision surface mount technology (SMT) equipment widely used in the electronic industry, a chip appearance defect-detection method based on multi-order fractional discrete wavelet packet decomposition (DWPD) is proposed in this paper. First, lead and body regions were extracted from chip images using the image segmentation algorithm with asymmetric Laplace mixture model and connected-component labelling algorithm; then, the texture feature of the region to be inspected was extracted with the multi-order fractional DWPD algorithm and the geometric and gradient features were combined to form image features of the region to be inspected before the subset of features was selected from image features with the feature selection algorithm based on the variational Bayesian Gaussian mixture model; and finally, the support vector machine was used to determine whether the region to be inspected was defective. An experiment was conducted on a data set captured in high-precision SMT equipment. The accuracy of the proposed chip appearance defect-detection method is about 93%, which is more accurate than existing ones.

How to apply O3 and PM2.5 collaborative control to practical management in China: A study based on meta-analysis and machine learning

Constant increase of atmospheric O3 concentration is a barrier for the further air quality improvement in China. Given that PM2.5 is still controlled as a key pollutant, managements for the collaborative reduction of O3 and PM2.5 are urgently required in China. In the current work, monitoring data of O3 and PM2.5 from 2015 to 2016 in 1464 monitoring sites (MS) was collected and cleaned. Additionally, 7 anthropogenic emission reductions were jointed with the corresponding monitoring data. According to the O3 and PM2.5 variation, a meta-analysis was conducted and divided regions into 4 categories via the effect size, region I: O3 and PM2.5 collaborative reduction, region II: PM2.5 decreased and O3 increased, region III: O3 decreased and PM2.5 increased, regions IV: both O3 and PM2.5 increased. Then, based on the region labels, machine learning was used to identify the pattern between region label and its precursor reductions. The findings were as follows: (1) Principal component analysis showed that NH3 was not focused on. (2) Random forest had a well performance on region classification with the accuracy of 80.40% and the importance of the 7 precursors was in the sequence of VOCs>NH3 > PM2.5 > OC > SO2 > NOX > Coarse PM. (3) Polytomous logistic regression evaluated the critical factors that influenced the region label, which showed that the reductions of VOCs, NH3 and PM2.5 could achieve the collaborative reduction in a short time in most of cities in China. Based on the statistical results above, a kinetic management system including evaluation and policy-making sections was finally established, which filled the gap of the collaborative reduction in environmental management in China.

Automatic detection of brain metastases on contrast-enhanced CT with deep-learning feature-fused single-shot detectors

PurposeDespite the potential usefulness, no automatic detector is available for brain metastases on contrast-enhanced CT (CECT). The study aims to develop and investigate deep learning–based detectors for brain metastases detection on CECT. MethodThe study included 127 CECTs from 127 patients (65.5 years±11.1; 87 men). The ground-truth annotation was performed semi-automatically by applying connected-component analysis to the binarized dataset by three radiologists. Single-shot detector (SSD) algorithms, with and without a feature-fusion module, were developed and trained using 97 scans. The performance was evaluated at the detector’s 50 % confidence threshold with the remaining 30 scans using sensitivity, positive-predictive value (PPV), and the false-positive rate per scan (FPR). ResultsFeature-fused SSD achieved an overall sensitivity of 88.1 % (95 % confidence interval [CI]: [85.2 %,88.6 %]; 214/243) and PPV of 36.0 % (95 % CI: [33.7 %,37.1 %]; 233/648), with 13.8 FPR (95 % CI: [12.7,15.0]). Lesions < 3 mm had a sensitivity of 23.1 % (95 % CI: [21.2 %,40.0 %]; 3/13), with 0.2 FPR (95 % CI: [0.23,0.65]). Lesions measuring 3–6 mm had a sensitivity of 80.0 % (95 % CI: [76.0 %,79.8 %]); 60/75) with 5.8 FPR (95 % CI: [5.0,6.2]). Lesions > 6 mm had a sensitivity of 97.4 % (95 % CI: [94.1 %,97.4 %]); 151/155) with 7.9 FPR (95 % CI: [7.2,8.5]). Feature-fused SSD had a significantly higher overall sensitivity (p = 0.03, t = 2.75) or sensitivity for lesions < 3 mm (p = 0.002, t = 4.49) than baseline SSD, while the overall PPV was similar (p = 0.96, t = -0.02). ConclusionsThe SSD algorithm identified brain metastases on CECT with reasonable accuracy for lesions > 3 mm without pre/post-processing.

Improved copy move forgery detection method via L*a*b* color space and enhanced localization technique

The wide availability of easy-to-use image editors has made the authenticity of images questionable. Copy-move is one of the most applied forgery types. A new copy-move forgery detection and localization technique independent from the characteristics of the forged regions is proposed in this paper. SIFT keypoints are obtained from CLAHE applied sub-images extracted from the input image by using RGB and L*a*b* color-spaces. Keypoint matching is realized on the sub-images and duplicated regions are determined roughly to create roughly marked image R. RANSAC is also applied in this stage and generated homography matrix is used to construct transformed roughly marked image R′. The method extracts DCT based features from R and R′ to localize exact borders of the tampered regions on the roughly determined areas by using a dynamic threshold. The proposed method has a new suggestion to determine the threshold dynamically. Tamper localization procedure also utilizes from morphological operations (chosen depending on the characteristic of the image) and Connected Component Labeling to determine exact forge boundaries. Results indicate that the proposed method has a better performance compared with state-of-the-art copy-move forgery detection methods on the GRIP dataset. Scaling attack performance of the method is especially better than similar works as shown in the results.

Steerable needles for radio-frequency ablation in cirrhotic livers

Accurate needle placement in deep-seated liver tumours can be difficult. In this work, we disclose two new manually controlled steerable needles for 17G radio-frequency ablation probe placement. The needles contain stylets with embedded compliant joints for active tip articulations, and concentric tubes for (curved-path) guidance. Needle steering was evaluated sequentially by intended users and in intended-use tissue types. Six interventional radiologists evaluated the needle in repeated ultrasound-guided steering tasks in liver-mimicking phantoms. Targets were located at a 100 mm depth and 20 mm lateral offset from the initial insertion line. The resulting mean absolute tip placement error was 1.0 ± 1.0 mm. Subsequently, steering-induced tissue damage was evaluated in fresh cirrhotic human liver explants. The surface area of puncture holes was estimated in scanned histology slides, using a connected-components analysis. The mean surface area was 0.26 ± 0.16 mm2 after steering with a median radius of curvature of 0.7 × 103 mm, versus 0.35 ± 0.15 mm2 after straight-path insertions with the steerable needle and 0.15 ± 0.09 mm2 after straight-path RFA probe insertions. The steering mechanisms proposed enable clinically relevant path corrections for 17G needles. Radiologists were quickly adept in curved-path RFA probe placement and the evaluation of histological tissue damage demonstrated a potentially safe use during liver interventions.

MCNet: Multi-level Correction Network for thermal image semantic segmentation of nighttime driving scene

Current state-of-the-art image segmentation methods are mainly based on visible spectrum images. However, it remains challenging under adverse environmental conditions (e.g., nighttime, foggy, snowy). Thermal images can be captured in darkness and deployed in any weather and time. But the research focus on thermal images semantic segmentation remains underexploited since the lack of thermal images segmentation dataset of nighttime driving scene. On the other hand, existing methods do not adequately capture context between related pixels and edge details in thermal images. In this work, we focus on thermal image semantic segmentation of nighttime driving scene. First, we propose a multi-level correction network (MCNet) with a multi-level attention module (MAM) and a multi-level edge enhancement module (MEEM). Specifically, MAM selectively captures the inter-class and intra-class contextual dependencies by a multi-level correction process. In addition, MEEM continuously combines precise context information and edge prior knowledge in each level to correct the final feature representation. Second, we present a new dataset called SCUT-Seg which contains 2010 thermal images recorded from various road scenes with 10 manually annotated semantic region labels. Extensive experiments demonstrate that the proposed method performs favorably against the state-of-the-art methods on SCUT-Seg and the public MFNet dataset. The source code and dataset are available at https://github.com/haitaobiyao/MCNet.

Automated filtering and polygonal reconstruction of virtual skull model based on computed tomography

A method of automated filtering and polygonal reconstruction of the skull virtual model using the methods of “Marching cubes” and “Connected component labeling” is considered. An algorithm for effective software implementation is described. A computational experiment on computed tomography data consisting of 244 DICOM images is presented. The experimental results confirm the correctness and computational efficiency of the algorithm.

One DAG to Rule Them All.

In this paper, we present novel strategies for optimizing the performance of many binary image processing algorithms. These strategies are collected in an open-source framework, GRAPHGEN, that is able to automatically generate optimized C++ source code implementing the desired optimizations. Simply starting from a set of rules, the algorithms introduced with the GRAPHGEN framework can generate decision trees with minimum average path-length, possibly considering image pattern frequencies, apply state prediction and code compression by the use of Directed Rooted Acyclic Graphs (DRAGs). Moreover, the proposed algorithmic solutions allow to combine different optimization techniques and significantly improve performance. Our proposal is showcased on three classical and widely employed algorithms (namely Connected Components Labeling, Thinning, and Contour Tracing). When compared to existing approaches -in 2D and 3D-, implementations using the generated optimal DRAGs perform significantly better than previous state-of-the-art algorithms, both on CPU and GPU.

Non-Hodgkin Type Lymphoma Cancer Cell Detection using Connected Components Labeling and Moments of Image

Cancers are one of the deadliest diseases with a costly treatment system in the world at present. In this paper a cost-effective, autonomous system of cancer-cell detection was proposed using several efficient image processing methods to develop an early stage non-Hodgkin type lymphoma which is a type of blood cancer. The system is implemented automatically to detect the traits of cancer in microscopy images of biopsy samples. Recent attempts have previously lacked flexibility in characteristics and the accuracy level is not consistent with the individual cancer type. The framework consisted three stages for detecting cancer on the basis of various detected traits including cell segmentation, quantification, area measurement analysis of cells, a center clump detection using the moment of image, identification of 4-connected components and Moore-Neighbor tracing algorithm. This methodology has been used in several sets of images and Feedback from these test executions has been used to improve the system. Subsequently, the proposed method can be used efficiently for used for autonomous non-hodgking type lymphoma cancer cell detection, which has an accuracy of 93.75%.

Dental Images' Segmentation Using Threshold Connected Component Analysis.

Recent advances in medical imaging analysis, especially the use of deep learning, are helping to identify, detect, classify, and quantify patterns in radiographs. At the center of these advances is the ability to explore hierarchical feature representations learned from data. Deep learning is invaluably becoming the most sought out technique, leading to enhanced performance in analysis of medical applications and systems. Deep learning techniques have achieved great performance results in dental image segmentation. Segmentation of dental radiographs is a crucial step that helps the dentist to diagnose dental caries. The performance of these deep networks is however restrained by various challenging features of dental carious lesions. Segmentation of dental images becomes difficult due to a vast variety in topologies, intricacies of medical structures, and poor image qualities caused by conditions such as low contrast, noise, irregular, and fuzzy edges borders, which result in unsuccessful segmentation. The dental segmentation method used is based on thresholding and connected component analysis. Images are preprocessed using the Gaussian blur filter to remove noise and corrupted pixels. Images are then enhanced using erosion and dilation morphology operations. Finally, segmentation is done through thresholding, and connected components are identified to extract the Region of Interest (ROI) of the teeth. The method was evaluated on an augmented dataset of 11,114 dental images. It was trained with 10 090 training set images and tested on 1024 testing set images. The proposed method gave results of 93% for both precision and recall values, respectively.