Adaptive Descriptor Research Articles

Few-shot anomaly detection with adaptive feature transformation and descriptor construction

Anomaly Detection (AD) has been extensively adopted in industrial settings to facilitate quality control of products. It is critical to industrial production, especially to areas such as aircraft manufacturing, which require strict part qualification rates. Although being more efficient and practical, few-shot AD has not been well explored. The existing AD methods only extract features in a single frequency while defects exist in multiple frequency domains. Moreover, current methods have not fully leveraged the few-shot support samples to extract input-related normal patterns. To address these issues, we propose an industrial few-shot AD method, Feature Extender for Anomaly Detection (FEAD), which extracts normal patterns in multiple frequency domains from few-shot samples under the guidance of the input sample. Firstly, to achieve better coverage of normal patterns in the input sample, we introduce a Sample-Conditioned Transformation Module (SCTM), which transforms support features under the guidance of the input sample to obtain extra normal patterns. Secondly, to effectively distinguish and localize anomaly patterns in multiple frequency domains, we devise an Adaptive Descriptor Construction Module (ADCM) to build and select pattern descriptors in a series of frequencies adaptively. Finally, an auxiliary task for SCTM is designed to ensure the diversity of transformations and include more normal patterns into support features. Extensive experiments on two widely used industrial AD datasets (MVTec-AD and VisA) demonstrate the effectiveness of the proposed FEAD.

Formation containment control of multi‐agent systems with actuator and sensor faults

AbstractThis article investigates the fault‐tolerant formation containment control of multi‐agent systems with actuator and sensor faults. First, adaptive descriptor observers are proposed for the followers to estimate system states, actuator faults, and sensor faults simultaneously. Then, distributed fault‐tolerant formation containment controllers are designed based on the estimated values and sufficient conditions for the formation containment control are identified using Lyapunov method in the time domain. Finally, the effectiveness and advantages of the proposed method are demonstrated through numerical simulations.

AdaSG: A Lightweight Feature Point Matching Method Using Adaptive Descriptor with GNN for VSLAM.

Feature point matching is a key component in visual simultaneous localization and mapping (VSLAM). Recently, the neural network has been employed in the feature point matching to improve matching performance. Among the state-of-the-art feature point matching methods, the SuperGlue is one of the top methods and ranked the first in the CVPR 2020 workshop on image matching. However, this method utilizes graph neural network (GNN), resulting in large computational complexity, which makes it unsuitable for resource-constrained devices, such as robots and mobile phones. In this work, we propose a lightweight feature point matching method based on the SuperGlue (named as AdaSG). Compared to the SuperGlue, the AdaSG adaptively adjusts its operating architecture according to the similarity of input image pair to reduce the computational complexity while achieving high matching performance. The proposed method has been evaluated through the commonly used datasets, including indoor and outdoor environments. Compared with several state-of-the-art feature point matching methods, the proposed method achieves significantly less runtime (up to 43× for indoor and up to 6× for outdoor) with similar or better matching performance. It is suitable for feature point matching in resource constrained devices.

Phase Congruency Order-Based Local Structural Feature for SAR and Optical Image Matching

Automatic matching of synthetic aperture radar (SAR) and optical images is a fundamental task in many remote sensing applications. However, due to different imaging modalities, conventional matching methods provide limited performances. In this letter, based on the observation that structural features are maintained across different modality images, we propose a novel feature-based method to effectively address SAR and optical image matching. The proposed method is built on the phase congruency (PC) model and consists mainly of two stages. First, a modified version of the uniform nonlinear diffusion-based Harris (MUND-Harris) detector is introduced to extract the local features. Unlike the UND-Harris, MUND-Harris employs the PC instead of image intensity for feature extraction and thus obtain well-distributed and highly repeatable feature points. Second, a local structural descriptor, namely phase congruency order-based local structural (PCOLS), is designed for the extracted points. PCOLS is constructed in a grouping manner and further encodes image structures with an adaptive descriptor structure, which provides robustness against modality variations including significant geometric and intensity differences. Experiment results obtained on several SAR and optical image pairs demonstrate the encouraging performance of the proposed method.

A Framework for Instantaneous Driver Drowsiness Detection Based on Improved HOG Features and Naïve Bayesian Classification.

Due to their high distinctiveness, robustness to illumination and simple computation, Histogram of Oriented Gradient (HOG) features have attracted much attention and achieved remarkable success in many computer vision tasks. In this paper, an innovative framework for driver drowsiness detection is proposed, where an adaptive descriptor that possesses the virtue of distinctiveness, robustness and compactness is formed from an improved version of HOG features based on binarized histograms of shifted orientations. The final HOG descriptor generated from binarized HOG features is fed to the trained Naïve Bayes (NB) classifier to make the final driver drowsiness determination. Experimental results on the publicly available NTHU-DDD dataset verify that the proposed framework has the potential to be a strong contender for several state-of-the-art baselines, by achieving a competitive detection accuracy of 85.62%, without loss of efficiency or stability.

Fault estimation and tolerant control for discrete‐time nonlinear stochastic multiple‐delayed systems with intermittent sensor and actuator faults

SummaryThis article is concerned with the design of the observer and controller to perform fault estimation (FE) and fault tolerant control for the discrete‐time nonlinear stochastic systems with multiple time‐varying state delays and intermittent sensor and actuator faults. In comparison with the present research, the method addressed in this article is more flexible and feasible. A novel adaptive descriptor observer is developed to obtain the error dynamics and then realize n‐step FE. Then, an active observer‐based controller is constructed to stabilize the closed‐loop system. Moreover, an augmented closed‐loop system is acquired and a series of delay‐dependent sufficient conditions are collected by the delay‐dependent Lyapunov function utilizing linear matrix inequalities. The method is less conservative compared with the ones of the existence. At last, a simulation example is presented to demonstrate the advantages and effectiveness of the approach proposed in the article.

Observer and fault‐tolerant controller design for discrete‐time multiple state‐delayed T–S fuzzy systems

This study is concerned with observer-based fault estimation (FE) and fault-tolerant controller design for a class of discrete-time Takagi–Sugeno (T–S) fuzzy systems. There exist multiple time-varying state delays, sensor and actuator faults, local non-linear dynamics and exogenous disturbances in the systems. In comparison with the existing results, the approach suggested in this study is more flexible and feasible. By means of the n-step induction FE, a novel fuzzy adaptive descriptor observer is developed to obtain the n-step error functions. Then, an active dynamic output feedback fault-tolerant controller is designed to stabilise the closed-loop fuzzy system. Furthermore, a set of delay-dependent sufficient conditions are provided by the fuzzy Lyapunov function which utilises the form of linear matrix inequalities. The stability results from the observer and the controller in this study have less conservatism compared with the ones from the existence of observers and fault-tolerant controllers. At last, a simulation example is presented to demonstrate the advantages and effectiveness of the approach proposed in the study.

Crop Images Segmentation using Adaptive Morphologic Descriptors

This research is focused on the segmentation improvement of crop images by using adaptive morphologic descriptors instead of classic algorithms like K-means and the top-hat operator using predefined shapes like disk or diamond. Obtained results shows that using an adaptive morphologic descriptor improves the segmentation performance against the classic shapes like disc and diamond. In order to measure the process a set of 60 crop images was used including their respective ground-truth images. The images were segmented using the K-Means algorithm and the top-hat operator with the disk and diamond shapes at different sizes into a range to validate their performance. In order to generate the adaptive morphologic descriptor, the Univariated Marginal Distribution Algorithm was used with no constraints by exploring a range of different sizes. Also, performance metrics like receiver operating characteristic and accuracy rate were applied to the generated data in order to assess the results.

Fault Estimation and Tolerant Control for Discrete-Time Multiple Delayed Fuzzy Stochastic Systems With Intermittent Sensor and Actuator Faults

This article is concerned with observer-based fault estimation (FE) and tolerant controller design for a series of discrete-time Takagi-Sugeno (T-S) fuzzy stochastic systems. There exist multiple time-varying state delays, intermittent sensor and actuator faults, nonlinear dynamics, and exogenous disturbances in the systems. Compared with the results of the existence, this approach suggested in this article is more flexible and feasible. By means of the FE information, a novel fuzzy adaptive descriptor observer is developed to obtain the error dynamics. Then, an active observer-based fault-tolerant controller is designed to stabilize the closed-loop fuzzy system. Furthermore, a set of delay-dependent sufficient conditions are provided by the fuzzy Lyapunov function with the way of linear matrix inequalities (LMIs), which has less conservatism compared with the ones of the existing observers and fault-tolerant controllers. Finally, a simulation example is shown to illustrate the advantages and effectiveness of this approach depicted in this article.

Estimation using L1 adaptive descriptor observer for multivariable systems with nonlinear uncertainties and measurement noises

In this paper, an L1 adaptive descriptor observer is designed for multivariable systems with nonlinear uncertainties and measurement noises. If the system is detectable, noises are bounded and some rank conditions are satisfied, an L1 adaptive descriptor observer is constructed to asymptotically estimate states, nonlinear uncertainties and measurement noises at the same time. The original system is augmented with all the system states and measurement noises, two design parameters provide additional degrees of freedom. The freedom of selecting these parameters allows us to choose the derivative gain to reduce the noise amplification, the proportional gain to ensure the stability of the estimated error dynamics. An adaptive law will update the adaptive parameters which represent the uncertainty estimates such that the estimation error between the predicted state and the real state is driven to zero at every integration time-step. Of course, neglection of the unknowns for solving the error dynamic equations will introduce an estimation error in the adaptive parameters. The magnitude of this error can be lessened by choosing the time step as small as possible, meanwhile, the picking up of time step should satisfy the hardware requirement. The two design parameters and adaptive law guarantee the performance bounds for the estimation errors, both states and nonlinear uncertainties. Numerical examples are given to illustrate the design procedures, and the simulation results demonstrate the satisfactory tracking performance.

Intensity guided cost metric for fast stereo matching under radiometric variations.

Reliable and efficient stereo matching is a challenging task due to the presence of multiple radiometric variations. In stereo matching, correspondence between left and right images can become hard owing to low correlation between radiometric changes in left and right images. Previously presented cost metrics are not robust enough against intensive radiometric variations and/or are computationally expensive. In this work, we propose a new similarity metric coined as Intensity Guided Cost Metric (IGCM). IGCM turns out to significantly contribute to the depth accuracy by rejecting outliers and reducing the edge-fattening effect in object boundaries. IGCM is further combined explicitly with a color formation model to handle various radiometric changes that occur between stereo images. Experimental results on Middlebury dataset show 13.8%, 22.8%, 20.9%, 19.5 % and 9.1% decrease in average error rate compared to Adaptive Normalized Cross-Correlation (ANCC), Dense Adaptive Self-Correlation (DASC), Adaptive Descriptor(AD), Fast Cost Volume Filtering (FCVF) and Iterative Guided Filter (IGF)-based methods, respectively. Moreover, using integral images IGCM can achieve a speedup of 20x, 6x, 41x, 25x and 45x compared to the aforementioned methods.

Fault-Tolerant Control of a Nonlinear System Based on Generalized Fuzzy Hyperbolic Model and Adaptive Disturbance Observer

This paper focuses on the problem of fault-tolerant control for a class of nonlinear systems with unknown nonlinear dynamic. The generalized fuzzy hyperbolic model is used to approximate the unknown nonlinear function. A novel fuzzy adaptive descriptor observer and an adaptive disturbance observer are designed simultaneously to estimate the system state, actuator fault, sensor fault, and disturbance. In this paper, the disturbance is assumed to be generated from an unknown exogenous system. Compared with the existing results, in which the disturbance is generated from known exogenous system, the proposed adaptive disturbance observer is more applicable in practice, since the exogenous system is independent of the controlled system and its parameter matrix is difficult to be measured. Furthermore, using the estimation information, a novel nonlinear dynamic output feedback controller is designed. The controller can stabilize the nonlinear system, which is subjected to the unknown nonlinear dynamic, actuator fault, sensor fault, and mismatched disturbance, simultaneously. Finally, simulation results are provided to illustrate the effectiveness of the proposed techniques.

Detecting and inferring repetitive elements with accurate locations and shapes from façades

The use of repetition detection is an effective approach for increasing the efficiency of urban modeling. In practice, repetition detection can benefit from the apparent regularities and strong contextual relationships in facades. In view of this, we propose a novel algorithm for automatically detecting and inferring repetitive elements with accurate locations and shapes from facades. More specifically, firstly, starting from a rectification of the input facade, we employ the color clustering method to automatically derive candidate templates. Secondly, to detect the non- and partially occluded repetitive elements matching with the derived templates, we construct an adaptive region descriptor and a repetitive characteristic curve. Finally, the fully occluded elements are inferred by utilizing the Bayesian probability network, which can be learned from a database of the selected facades. The accuracy of our detection and inference is tested through a variety of experiments, and all of them justify the robustness of our algorithm to outliers such as appearance variations and occlusions.

Adaptive descriptor based on the geometric consistency of local image features: application to flower image classification

Geometric consistency is, usually, considered as a postprocessing step to filter matched sets of local features in order to discard outliers. In this work, it is used to propose an adaptive feature that describes the geometric dispersion of keypoints. It is based on a distribution computed by a nonparametric estimator so that no assumption is made about the data. We investigated and discussed the invariance properties of our descriptor under the most common two- and three-dimensional transformations. Then, we applied it to flower recognition. The classification is performed using the precomputed kernel of support vector machines classifier. Indeed, a similarity computing framework that uses the Kullback–Leibler divergence is presented. Furthermore, a customized layout for each flower image is designed to describe and compare separately the boundary and the central area of flowers. Experimentations made on the Oxford flower-17 dataset prove the efficiency of our method in terms of classification accuracy and computational complexity. The limits of our descriptor are also discussed on a 10-class subset of the Oxford flower-102 dataset.

Disaster Damage Assessment of Buildings Using Adaptive Self-Similarity Descriptor

Assessment of damage caused by a disaster is significant for coordinating emergency response teams and planning emergency aid. In this letter, a robust method for rapid building damage assessment is proposed using pre- and postevent EO images and building footprints. The method uses a local self-similarity descriptor (SSD) for change detection in buildings, which is shown to be robust against variations in global illumination and small local deformations. The use of building footprints helps reduce the false alarms due to changes in nonbuilding areas. Footprint is also used to differentiate small and large buildings, extract the boundary region of a building, and adapt the descriptor computation accordingly. It is shown that the adaptive SSD provides a more accurate measure of local damage on the building. The 2010 Haiti Earthquake and Typhoon Haiyan 2013 Philippines are analyzed with the proposed method, and 75/82% true positive rate and 25/15% false positive rate are obtained for detection of collapsed buildings with respect to the ground truth data of UNITAR/UNOSAT and HOT.

Exploring illumination robust descriptors for human epithelial type 2 cell classification

Strong illumination variation is a key challenge in the Human Epithelial Type 2 (HEp-2) cell classification task. Aiming to improve the robustness of the HEp-2 classification system to the illumination variation, this paper deeply explores discriminative and illumination robust descriptors. Specifically, we propose a novel Spatial Shape Index Descriptor (SSID) to capture spatial layout information of the second-order structures, and utilize a Local Orientation Adaptive Descriptor (LOAD), which was originally designed for texture classification, to the HEp-2 cell classification task. Both SSID and LOAD show strong discrimination and great complementarity to each other.Four different sets of experiments were carried out to evaluate SSID, LOAD and their combination. Our two submissions achieved superior performance on the new Executable Thematic of Pattern Recognition Techniques for Indirect Immunofluorescence images analysis. Compared to the rank 1st method in the ICPR 2014 HEp-2 cell classification contest, both of our submissions achieved a better performance when only using the provided training data. Our approaches also demonstrated superior performance on a newly compiled large-scale HEp-2 data set with 63,445 cell images.

Adaptive descriptor-based robust stereo matching under radiometric changes

In a real stereo vision system, the acquired stereo images suffer from varying radiometric changes due to illumination and camera parameter changes. Therefore, we propose an effective matching scheme created by building a content adaptive descriptor. Specifically, the descriptor reflects image contents and its element are adaptively weighted and applied to estimate the correct corresponding pixels based on the entropy energy function even under radiometric changes. For the performance evaluation, the proposed scheme is compared with the state-of-art algorithm using Middlebury and KITTI Vision stereo datasets that have radiometric changes. Specifically, 24 of 71 indoor image pairs in the Middlebury and 3 of 7 outdoor pairs are selected, respectively. Experimental result shows that the proposed method reports 6.23% bad pixel matching on average, but it outperforms state-of-the-art algorithms by reducing around 2% bad pixel matching error, which achieves about 16.5% performance improvement.

HEp-2 cell classification: The role of Gaussian Scale Space Theory as a pre-processing approach

Indirect Immunofluorescence Imaging of Human Epithelial Type 2 (HEp-2) cells is an effective way to identify the presence of Anti-Nuclear Antibody (ANA). Most existing works on HEp-2 cell classification mainly focus on feature extraction, feature encoding and classifier design. Very few efforts have been devoted to study the importance of the pre-processing techniques. In this paper, we analyze the importance of the pre-processing, and investigate the role of Gaussian Scale Space (GSS) theory as a pre-processing approach for the HEp-2 cell classification task. We validate the GSS pre-processing under the Local Binary Pattern (LBP) and the Bag-of-Words (BoW) frameworks. Under the BoW framework, the introduced pre-processing approach, using only one Local Orientation Adaptive Descriptor (LOAD), achieved superior performance on the Executable Thematic on Pattern Recognition Techniques for Indirect Immunofluorescence (ET-PRT-IIF) image analysis. Our system, using only one feature, outperformed the winner of the ICPR 2014 contest that combined four types of features. Meanwhile, the proposed pre-processing method is not restricted to this work; it can be generalized to many existing works.

LOAD: Local orientation adaptive descriptor for texture and material classification

In this paper, we propose a novel local feature, called Local Orientation Adaptive Descriptor (LOAD), to capture regional texture in an image. In LOAD, we proposed to define point description on an Adaptive Coordinate System (ACS), adopt a binary sequence descriptor to capture relationships between one point and its neighbors and use multi-scale strategy to enhance the discriminative power of the descriptor. The proposed LOAD enjoys not only discriminative power to capture the texture information, but also has strong robustness to illumination variation and image rotation. Extensive experiments on benchmark data sets of texture classification and real-world material recognition show that the LOAD yields the state-of-the-art performance. It is worth to mention that we achieve a superior classification accuracy on Flickr Material Database by using a single feature. Moreover, by combining LOAD with Convolutional Neural Networks (CNN), we obtain significantly better performance than both the LOAD and CNN. This result confirms that the LOAD is complementary to the learning-based features.

Matching Images With Multiple Descriptors: An Unsupervised Approach for Locally Adaptive Descriptor Selection

With the aim to improve the performance of feature matching, we present an unsupervised approach for adaptive description selection in the space of homographies. Inspired by the observation that the homographies of correct feature correspondences vary smoothly along the spatial domain, our approach stands on the unsupervised nature of feature matching, and can choose a good descriptor locally for matching each feature point, instead of using one global descriptor. To this end, the homography space serves as the domain for selecting various heterogeneous descriptors. Correspondences obtained by any descriptors are considered as points in the space, and their geometric coherence and spatial continuity are measured via computing the geodesic distances. In this way, mutual verification across different descriptors is allowed, and correct correspondences will be highlighted with a high degree of consistency short geodesic distances here. It follows that one-class SVM can be applied to identifying these correct correspondences, and achieves adaptive descriptor selection. The proposed approach is comprehensively compared with the state-of-the-art approaches, and evaluated on five benchmarks of image matching. The promising results manifest its effectiveness.