Background Separation Research Articles

ANLPT: Self-Adaptive and Non-Local Patch-Tensor Model for Infrared Small Target Detection

Infrared small target detection is widely used for early warning, aircraft monitoring, ship monitoring, and so on, which requires the small target and its background to be represented and modeled effectively to achieve their complete separation. Low-rank sparse decomposition based on the structural features of infrared images has attracted much attention among many algorithms because of its good interpretability. Based on our study, we found some shortcomings in existing baseline methods, such as redundancy of constructing tensors and fixed compromising factors. A self-adaptive low-rank sparse tensor decomposition model for infrared dim small target detection is proposed in this paper. In this model, the entropy of image block is used for fast matching of non-local similar blocks to construct a better sparse tensor for small targets. An adaptive strategy of low-rank sparse tensor decomposition is proposed for different background environments, which adaptively determines the weight coefficient to achieve effective separation of background and small targets in different background environments. Tensor robust principal component analysis (TRPCA) was applied to achieve low-rank sparse tensor decomposition to reconstruct small targets and their backgrounds separately. Sufficient experiments on the various types data sets show that the proposed method is competitive.

Weak-Strong Self-Adapting Fuzzy Neural Classifier for Dynamic Object Detection in RGBD Videos

This paper presents a fuzzy neural method to model background from videos in order to detect dynamic objects. The method includes a weak fuzzy classifier that performs an initial foreground and background separation based on color and depth differences between the actual frame and background models. The outputs of this fuzzy system are weighted according to the result of the color and depth noise modeling. A degree of uncertainty and the strength of decisions, in combination with the weighting results, are used by the method to define more accurately the dynamic objects through a strong fuzzy classifier. The final stage of foreground detection is implemented with a Discrete-Time Cellular Neural Network to improve the foreground definition. Finally, the color and depth background models are updated based on a fuzzy learning rate strategy. The method was evaluated with the new SBM-RGBD database and compared against several state-of-the-art methods showing a similar or better performance considering the quantitative and qualitative evaluations.

Tensor Spectral k-Support Norm Minimization for Detecting Infrared Dim and Small Target Against Urban Backgrounds

In the low-altitude urban background with heavy interference, especially in the face of corner interference with higher intensity than the target, infrared (IR) dim and small target is extremely lack of prior information (i.e., size, shape and contrast information). In such case, the existing detection methods usually suffer from high false alarm or even failure. To deal with this situation, we develop a novel spatial-temporal tensor model with tensor spectral <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">k</i> -support norm minimization (STTM-TSNM) for detecting IR dim and small target. Firstly, the spatial-temporal information of the original image sequence can be preserved completely by constructing the holistic STTM. Then, according to the spatial-temporal related prior knowledge of the target and background, the target detection task is customized as an optimization problem of low rank and sparse tensor recovery. To better preserve the internal structure and capture more global information, the tensor spectral <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">k</i> -support norm minimization is introduced as the regularization term of the constraint background. Finally, draw support from the framework of alternating direction method of multipliers (ADMM) algorithm, the precise separation of target and background is achieved. In addition, to promote the prosperity of sequential detection methods, we released to the scientific community a small IR target dataset containing six image sequences with urban background. The experimental results on six real IR sequences demonstrate that our method outputs the most outstanding detection performance compared with the latest sequential detection methods.

Light Field Reflection and Background Separation Network Based on Adaptive Focus Selection

Separating the background and the reflection layer is an ill-posed but meaningful task. Since the two layers often have different depths, Light Fields (LFs), which contain the depth information, show great potential to deal with this task. In this paper, we propose to fully exploit the difference between background and reflection layers from different kinds of images in LFs. Specifically, we fuse the complementary features from sub-aperture images, in which the overlaps of the two layers vary a lot. Then we propose the adaptive focus selection strategy and use the dynamic filter to select the appropriate focus. The refocused image is then obtained from the focal stack images, where the clearness of the two layers is different. Finally, the clean images are restored using a dual attention reconstruction module. Our model is the first one that successfully addresses the LF reflection separation task with an end-to-end deep neural network. The experiments show that the background and reflection layers are well separated, regardless of the relative intensity difference between the two layers on both synthetic and real-world datasets. The quantitative and qualitative comparison results show that our method achieves better performance than other state-of-the-art methods.

All-optical quantum vacuum signals in two-beam collisions

We study the collision of two optical laser pulses in a pump-probe setup using beams with circular and elliptic cross section and estimate the number of discernible signal photons induced by quantum vacuum nonlinearities. In this analysis we study strategies to optimize the quantum vacuum signal discernible from the background of the driving lasers. One of the main results is that the collision of two maximally focused lasers does not lead to the best discernible signal. Instead, widening the focus typically improves the signal to background separation in the far field. For petawatt class lasers, an optimal choice of the focus waist yields several discernible photons per shot in contrast to no discernible signal for tight focusing. Further enhancement is possible by using an elliptical waist.

Simultaneous Determination of Nicotine and Phenolic Compounds in Tobacco by Capillary Electrophoresis with Pipette Tip Electrodes

Aim: The aim of this work is to fabricate pipette tip electrodes for the capillary electrophoretic determination of nicotine and phenolic compounds in tobacco. Background: The content of nicotine affects not only the quality of tobacco products but also the health of smokers. Phenolic compounds are important flavor precursors in tobacco. The quantity of phenolic compounds is one of the most important evaluation indicators of tobacco quality. It is of high importance to determine nicotine and phenolic compounds in tobacco for quality control and the health of smokers. Objective: A method based on capillary electrophoresis and amperometric detection was developed for the simultaneous determination of nicotine, rutin, chlorogenic acid, quercetin, ferulic acid, gallic acid, and protocatechuic acid in tobacco leaves. Pipette electrodes were designed and fabricated for their amperometric detection. Method: Nicotine, rutin, chlorogenic acid, quercetin, ferulic acid, gallic acid, and protocatechuic acid were determined by capillary electrophoresis in combination with the detection electrodes that were fabricated by packing the composite of carbon nanotube and epoxy in pipette tips. Results: Detection potentials, the acidity and concentrations of background electrolyte, separation voltages, and injection times were optimized. At a high voltage of 12 kV, separation of the seven analytes could be achieved in less than 11 min in a piece of 40 cm long fused silica capillary with a background electrolyte of 50 mM borate buffer (9.2). Linearity was observed between the peak currents and the concentrations, with the limits of detection ranging from 0.1 to 0.2 μM for the seven analytes at the pipette electrodes. The method was applied in the simultaneous determination of nicotine and phenolic compounds with satisfactory assay results. Conclusions: The pipette tip electrodes were successfully coupled with capillary electrophoresis for tobacco analysis. The CE-AD method provides not only a simple approach for the quality control of tobacco and its preparations but also an alternative technique for the constituent and fingerprint investigation of other plants.

Kernel Minimum Noise Fraction Transformation-Based Background Separation Model for Hyperspectral Anomaly Detection

A significant challenge in methods for anomaly detection (AD) in hyperspectral images (HSIs) is determining how to construct an efficient representation for anomalies and background information. Considering the high-order structures of HSIs and the estimation of anomalies and background information in AD, this article proposes a kernel minimum noise fraction transformation-based background separation model (KMNF-BSM) to separate the anomalies and background information. First, spectral-domain KMNF transformation is performed on the original hyperspectral data to fully mine the high-order correlation between spectral bands. Then, a BSM that combines the outlier removal, the iteration strategy, and the Reed–Xiaoli detector (RXD) is proposed to obtain accurate anomalous and background pixel sets based on the extracted features. Finally, the anomalous and background pixel sets are used as input for anomaly detectors to improve the background suppression and anomaly detection capabilities. Experiments on several HSIs with different spatial and spectral resolutions over different scenes are performed. The results demonstrate that the KMNF-BSM-based algorithms have better target detectability and background suppressibility than other state-of-the-art algorithms.

Unsupervised Class-Agnostic Instance Segmentation of 3D LiDAR Data for Autonomous Vehicles

Fine-grained scene understanding is essential for autonomous driving. The context around a vehicle can change drastically while navigating, making it hard to identify and understand the different objects that may appear. Although recent efforts on semantic and panoptic segmentation pushed the field of scene understanding forward, it is still a challenging task. Current methods depend on annotations provided before deployment and are bound by the labeled classes, ignoring long-tailed classes not annotated in the training data due to the scarcity of examples. However, those long-tailed classes, such as baby strollers or unknown animals, can be crucial when interpreting the vehicle surroundings, e.g., for safe interaction. We address the problem of class-agnostic instance segmentation in this paper that also tackles the long-tailed classes. We propose a novel approach and a benchmark for class-agnostic instance segmentation and a thorough evaluation of our method on real-world data. Our method relies on a self-supervised trained network to extract point-wise features to build a graph representation of the point cloud. Then, we use GraphCut to perform foreground and background separation, achieving instance segmentation without requiring any label. Our results show that our approach is able to achieve instance segmentation and a competitive performance compared to state-of-the-art supervised methods.

Custom Orthogonal Weight functions (COWs) for event classification

A common problem in data analysis is the separation of signal and background. We revisit and generalise the so-called sWeights method, which allows one to calculate an empirical estimate of the signal density of a control variable using a fit of a mixed signal and background model to a discriminating variable. We show that sWeights are a special case of a larger class of Custom Orthogonal Weight functions (COWs), which can be applied to a more general class of problems in which the discriminating and control variables are not necessarily independent and still achieve close to optimal performance. We also investigate the properties of parameters estimated from fits of statistical models to sWeighted data and provide closed formulas for the asymptotic covariance matrix of the fitted parameters. To illustrate our findings, we discuss several practical applications of these techniques.

Moving Object Detection in Dynamic Environment via Weighted Low-Rank Structured Sparse RPCA and Kalman Filtering

As a classical problem for computer vision, moving object detection (MOD) can be efficiently achieved by foreground and background separation. The Robust Principal Component Analysis (RPCA)-based method has been potentially utilized to solve the problem. However, the detection accuracy for RPCA-based method is limited for complex scenes with slow-motion. Besides, it is time consuming for the way to seek for background modeling based on solving a low-rank minimization problem, for which multiple frames of the videos are required as the input. Therefore, a real-time MOD framework (LSRPCA_KF) is proposed for the dynamic background, where a weighted low-rank and structured sparse RPCA algorithm is used to achieve background modeling for history data, while the online MOD is achieved by the background subtraction method and updated by the Kalman filter for every real time frame. Specifically, for the background model, a newly designed weight is incorporated to distinguish the significance of different singular values, and a structured sparse prior is added to penalty the spatial connection property of the moving object. Besides, the weighted low-rank and structured sparse RPCA model is efficiently solved by the Alternating Direction Method of Multipliers (ADMM) optimization algorithm. Experimental results demonstrated that better performance of our method with significantly reduced delay in processing and better detect the moving object has been achieved, especially for the dynamic background.

A Random CNN Sees Objects: One Inductive Bias of CNN and Its Applications

This paper starts by revealing a surprising finding: without any learning, a randomly initialized CNN can localize objects surprisingly well. That is, a CNN has an inductive bias to naturally focus on objects, named as Tobias ("The object is at sight") in this paper. This empirical inductive bias is further analyzed and successfully applied to self-supervised learning (SSL). A CNN is encouraged to learn representations that focus on the foreground object, by transforming every image into various versions with different backgrounds, where the foreground and background separation is guided by Tobias. Experimental results show that the proposed Tobias significantly improves downstream tasks, especially for object detection. This paper also shows that Tobias has consistent improvements on training sets of different sizes, and is more resilient to changes in image augmentations.

A New Method for Inverter Diagnosis of Electric Locomotive Using Adversarial Neural Networks

In order to improve the fault diagnosis accuracy of the electric locomotive inverter, this article combines the adversarial neural network to construct the electric locomotive inverter diagnosis system. Moreover, at the data level, this article compares and analyzes three methods of data expansion based on single-sample processing, data expansion based on image front and background separation, and data expansion based on an adversarial neural network. In addition, this article adopts a new feature extractor and increases the penalty cost of small samples being misclassified. Finally, this article uses the LBP operator to extract the image texture features to distinguish and detect the different shapes of the rotor windings and build an intelligent system to verify the effect of the proposed system model. The experimental research shows that the inverter diagnosis system for electric locomotives based on the proposed adversarial neural network has a good practical effect.

Home Environment Augmented Reality System Based on 3D Reconstruction of a Single Furniture Picture.

With the popularization of the concept of “metaverse”, Augmented Reality (AR) technology is slowly being applied to people’s daily life as its underlying technology support. In recent years, rapid 3D reconstruction of interior furniture to meet AR shopping needs has become a new method. In this paper, a virtual home environment system is designed and the related core technologies in the system are studied. Background removal and instance segmentation are performed for furniture images containing complex backgrounds, and a Bayesian Classifier and GrabCut (BCGC) algorithm is proposed to improve on the traditional foreground background separation technique. The reconstruction part takes the classical occupancy network reconstruction algorithm as the network basis and proposes a precise occupancy network (PONet) algorithm, which can reconstruct the structural details of furniture images, and the model accuracy is improved. Because the traditional 3D registration model is prone to the problems of model position shift and inaccurate matching with the scene, the AKAZE-based tracking registration algorithm is improved, and a Multiple Filtering-AKAZE (MF-AKAZE) based on AKAZE is proposed to remove the matching points. The matching accuracy is increased by improving the RANSAC filtering mis-matching algorithm based on further screening of the matching results. Finally, the system is verified to realize the function of the AR visualization furniture model, which can better complete the reconstruction as well as registration effect.

A novel hand gesture detection and recognition system based on ensemble-based convolutional neural network

Nowadays, hand gesture recognition has become an alternative for human-machine interaction. It has covered a large area of applications like 3D game technology, sign language interpreting, VR (virtual reality) environment, and robotics. But detection of the hand portion has become a challenging task in computer vision and pattern recognition communities. Deep learning algorithm like convolutional neural network (CNN) architecture has become a very popular choice for classification tasks, but CNN architectures suffer from some problems like high variance during prediction, overfitting problem and also prediction errors. To overcome these problems, an ensemble of CNN-based approaches is presented in this paper. Firstly, the gesture portion is detected by using the background separation method based on binary thresholding. After that, the contour portion is extracted, and the hand region is segmented. Then, the images have been resized and fed into three individual CNN models to train them in parallel. In the last part, the output scores of CNN models are averaged to construct an optimal ensemble model for the final prediction. Two publicly available datasets (labeled as Dataset-1 and Dataset-2) containing infrared images and one self-constructed dataset have been used to validate the proposed system. Experimental results are compared with the existing state-of-the-art approaches, and it is observed that our proposed ensemble model outperforms other existing proposed methods.

Monitoring of amoxicilline and ceftazidime in the microdialysate of diabetic foot and serum by capillary electrophoresis with contactless conductivity detection.

Determination of the broad-spectrum antibiotics amoxicilline (AMX) and ceftazidime (CTZ) in blood serum and microdialysates of the subcutaneous tissue of the lower limbs is performed using CE withcontactless conductivity detection (C4 D). Baseline separation of AMX is achieved in 0.5 M acetic acid as the background electrolyte and separation of CTZ in3.2 M acetic acid with addition of 13% v/v methanol. The CE-C4 D determination is performed in a 25µm capillary with suppression of the EOF using INST-coating on an effective length of 18cm and the attained migration time is 4.2min for AMX and 4.4min for CTZ. The analysis was performed using 20µl of serum and 15µl of microdialysate, treated by the addition of acetonitrile in a ratio of 1/3 v/v and the sample is injected into the capillary using the large volume sample stacking technique. The LOQ attained in the microdialysate is 148ng/ml for AMX and 339ng/ml for CTZ, and in serum 143ng/ml for AMX and 318ng/ml for CTZ. The CE-C4 D method is employed for monitoring the passage of AMX and CTZ from the blood circulatory system into the subcutaneous tissue at the sites of diabetic ulceration in patients suffering from diabetic foot syndrome and also for measuring the pharmacokinetics following intravenous application of bolus antibiotic doses.

Power Monitoring Image Encryption Transmission Method Based on Internet of Things

In order to be able to achieve state maintenance, keep historical data of the operation of each system, improve the precision of computer room management, improve work efficiency, reduce the work intensity of computer room managers, and reduce manual errors and potential accidents, this paper combines the Internet of Things technology to construct an image encryption transmission method that can be used for power monitoring. Moreover, this paper constructs a power system intelligent monitoring system based on the Internet of Things, and encrypts relevant data. In addition, this paper proposes a surveillance video encryption algorithm based on foreground and background separation. This video encryption algorithm is different from ordinary video encryption methods. It fully taps the characteristics of the surveillance video taken by a still camera. According to the experimental research results, it can be seen that the power monitoring image encryption transmission system based on the Internet of Things constructed in this paper has good performance in power monitoring image processing, image encryption, image transmission, etc. The evaluation results of the three functional parameters are all above 80%, which have great advantages compared with traditional algorithms. Therefore, the method proposed in this paper has good practical effects.

The Relationship between Fe Mineralization and Magnetic Basement Faults using Multifractal Modeling in the Esfordi and Behabad Areas (BMD), Central Iran

AbstractMultifractal modeling is a mathematical method for the separation of a high potential mineralized background from a non‐mineralized background. The Concentration‐Distance to Fault structures (C‐DF) fractal model and the distribution of the known iron (Fe) deposits/mines seen in the Esfordi and Behabad 1:100,000 sheets from the Bafq region of central Iran are used to distinguish Fe mineralization based on their distance to magnetic basement structures and surface faults, separately, using airborne geophysical data and field surveys. Application of the C‐DF fractal model for the classification of Fe mineralizations in the Esfordi and Behabad areas reveals that the main ones show a correlation with their distance from magnetic basement structures. Accordingly, the distances of Fe mineralizations with grades of Fe higher than 55% (43% &lt; Fe ≤ 60%) are located at a distance of less than 1 km, whereas for surfacial faults with grades of 43% ≤ Fe ≤ 60%, the distances are 3162&lt; DF ≤ 4365 m from the faults. Thus, there is a positive relationship between Fe mineralization and magnetic basement structures. Also, the proximity evidence of Precambrian high‐grade Fe mineralization related to magnetic basement structures indicates syn‐rifting tectonic events. Finally, this C‐DF fractal model can be used for exploration of magmatic and hydrothermal ore deposits.

Detecting abnormality with separated foreground and background: Mutual Generative Adversarial Networks for video abnormal event detection

As one of the most important tasks in intelligent video analysis, video abnormal event detection has been extensively studied. Prior arts have made a great process in designing frameworks to capture spatio-temporal features of video frames. However, video frames usually consist of various objects. It is challenging to grasp the nuances of anomalies against noisy backgrounds. To tackle the bottleneck, we propose a novel Foreground–Background Separation Mutual Generative Adversarial Network (FSM-GAN) framework. The FSM-GAN permits the separation of video frames into the foreground and background. The separated foreground and background are utilized as the input of mutual generative adversarial networks, which transform raw-pixel images in optical-flow representations and vice versa. In the networks, the background is regarded as known conditions and the model focuses on learning the high-level spatio-temporal foreground features to represent the event with the given conditions during the mutual adversarial training. In the test stage, these high-level features instead of low-level visual primitives are utilized to measure the abnormality in the semantic level. Compared with state-of-the-art methods and other abnormal event detection approaches, the proposed framework demonstrates its effectiveness and reliability across various scenes and events.

Extraction and classification of apple defects under uneven illumination based on machine vision

AbstractThe extraction and classification of different kinds of defects can achieve a more sophisticated classification of apples. In this study, a region of interest (ROI) extraction algorithm based on background separation, brightness correction, and the global threshold was designed for the extraction of apple decay and bruise under uneven light conditions. After brightness correction, we calculate the average gray value of the apple region and multiply the value by 0.8 as the threshold. In this way, the threshold of different apple images can be determined automatically without a manual setting. After extracting ROI, we realized the classification of the two kinds of defects by extracting texture features. Based on the selected texture features, the support vector machine discriminant models were established. The accuracy of the model based on angular second moment was 94.8% and the model based on entropy was 94.7%. The total time from ROI extraction to classification was approximately 1.2 s. The results showed that this extraction and classification process had application prospects in practical detection classification.Practical ApplicationsUngraded fruits were not easy to store, transport, or sell at a price based on quality. The classification was directly related to fruit packaging, transportation, storage and sales of the results, and benefits, so it was very important to carry out fruit classification. Machine vision technology can achieve objective and nondestructive rapid detection. Many studies had achieved extraction of defective parts. However, they directly classified apples with defects into one category without further classification of types of the defects. In many classification standards, such as GBT10651‐2008, they need to classify apples more finely according to the defect types. This study achieved extraction and classification of different kinds of apple defects by machine vision technology and achieved a satisfactory accuracy and detection speed, which provided theoretical support for the realization of strict and fine classification of apple in industrial applications in the future.

A Method for Estimating the Background Column Concentration of CO2 Using the Lagrangian Approach

With the rapid growth of GHG monitoring satellites, more and more studies focused on the issue of inversion/optimization of CO2 fluxes using satellite-derived XCO2 observations in recent years. A common and critical challenge in this framework is the separation of background and anomalies from XCO2 observations, which directly affect performance of the CO2 fluxes inversion. We proposed a novel method to accurately extract background XCO2 from satellite observations. A series of observing system simulation experiments were performed to test the performance of the method. We found that the bias and uncertainty of the background concentration are below 0.01 ppm and 0.05 ppm in the given cases, respectively. Based on this method, we selected five overpasses from 2014 to 2016 to demonstrate a regional-scale flux inversion near Riyadh. The comparison with the two previous methods shows that the posterior simulated XCO2 by the method proposed in this paper can match better with the observed XCO2 from OCO-2.