Zhunga Liu

Haochen Chai, Meimei Su, Yang Lyu et al.

Fixed-wing Unmanned Aerial Vehicles (UAVs) are one of the most commonly used platforms for the burgeoning Low-altitude Economy (LAE) and Urban Air Mobility (UAM), due to their long endurance and high-speed capabilities. Classical obstacle avoidance systems, which rely on prior maps or sophisticated sensors, face limitations in unknown low-altitude environments and small UAV platforms. In response, this paper proposes a lightweight deep reinforcement learning (DRL) based UAV collision avoidance system that enables a fixed-wing UAV to avoid unknown obstacles at cruise speed over 30m/s, with only onboard visual sensors. The proposed system employs a single-frame image depth inference module with a streamlined network architecture to ensure real-time obstacle detection, optimized for edge computing devices. After that, a reinforcement learning controller with a novel reward function is designed to balance the target approach and flight trajectory smoothness, satisfying the specific dynamic constraints and stability requirements of a fixed-wing UAV platform. An adaptive entropy adjustment mechanism is introduced to mitigate the exploration-exploitation trade-off inherent in DRL, improving training convergence and obstacle avoidance success rates. Extensive software-in-the-loop and hardware-in-the-loop experiments demonstrate that the proposed framework outperforms other methods in obstacle avoidance efficiency and flight trajectory smoothness and confirm the feasibility of implementing the algorithm on edge devices. The source code is publicly available at \url{https://github.com/ch9397/FixedWing-MonoPPO}.

Wenyi Tan, Yang Li, Chenxing Zhao et al.

Object detection is a fundamental task in various applications ranging from autonomous driving to intelligent security systems. However, recognition of a person can be hindered when their clothing is decorated with carefully designed graffiti patterns, leading to the failure of object detection. To achieve greater attack potential against unknown black-box models, adversarial patches capable of affecting the outputs of multiple-object detection models are required. While ensemble models have proven effective, current research in the field of object detection typically focuses on the simple fusion of the outputs of all models, with limited attention being given to developing general adversarial patches that can function effectively in the physical world. In this paper, we introduce the concept of energy and treat the adversarial patches generation process as an optimization of the adversarial patches to minimize the total energy of the ``person'' category. Additionally, by adopting adversarial training, we construct a dynamically optimized ensemble model. During training, the weight parameters of the attacked target models are adjusted to find the balance point at which the generated adversarial patches can effectively attack all target models. We carried out six sets of comparative experiments and tested our algorithm on five mainstream object detection models. The adversarial patches generated by our algorithm can reduce the recognition accuracy of YOLOv2 and YOLOv3 to 13.19\% and 29.20\%, respectively. In addition, we conducted experiments to test the effectiveness of T-shirts covered with our adversarial patches in the physical world and could achieve that people are not recognized by the object detection model. Finally, leveraging the Grad-CAM tool, we explored the attack mechanism of adversarial patches from an energetic perspective.

Runqing Yang, Yimin Fu, Changyuan Wu et al.

Existing open set recognition (OSR) methods are typically designed for static scenarios, where models aim to classify known classes and identify unknown ones within fixed scopes. This deviates from the expectation that the model should incrementally identify newly emerging unknown classes from continuous data streams and acquire corresponding knowledge. In such evolving scenarios, the discriminability of OSR decision boundaries is hard to maintain due to restricted access to former training data, causing severe inter-class confusion. To solve this problem, we propose retentive angular representation learning (RARL) for incremental open set recognition (IOSR). In RARL, unknown representations are encouraged to align around inactive prototypes within an angular space constructed under the equiangular tight frame, thereby mitigating excessive representation drift during knowledge updates. Specifically, we adopt a virtual-intrinsic interactive (VII) training strategy, which compacts known representations by enforcing clear inter-class margins through boundary-proximal virtual classes. Furthermore, a stratified rectification strategy is designed to refine decision boundaries, mitigating representation bias and feature space distortion caused by imbalances between old/new and positive/negative class samples. We conduct thorough evaluations on CIFAR100 and TinyImageNet datasets and establish a new benchmark for IOSR. Experimental results across various task setups demonstrate that the proposed method achieves state-of-the-art performance.

Yimin Fu, Zhunga Liu, Dongxiu Guo et al.

The acquisition of high-quality labeled synthetic aperture radar (SAR) data is challenging due to the demanding requirement for expert knowledge. Consequently, the presence of unreliable noisy labels is unavoidable, which results in performance degradation of SAR automatic target recognition (ATR). Existing research on learning with noisy labels mainly focuses on image data. However, the non-intuitive visual characteristics of SAR data are insufficient to achieve noise-robust learning. To address this problem, we propose collaborative learning of scattering and deep features (CLSDF) for SAR ATR with noisy labels. Specifically, a multi-model feature fusion framework is designed to integrate scattering and deep features. The attributed scattering centers (ASCs) are treated as dynamic graph structure data, and the extracted physical characteristics effectively enrich the representation of deep image features. Then, the samples with clean and noisy labels are divided by modeling the loss distribution with multiple class-wise Gaussian Mixture Models (GMMs). Afterward, the semi-supervised learning of two divergent branches is conducted based on the data divided by each other. Moreover, a joint distribution alignment strategy is introduced to enhance the reliability of co-guessed labels. Extensive experiments have been done on the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset, and the results show that the proposed method can achieve state-of-the-art performance under different operating conditions with various label noises.

Lianmeng Jiao, Feng Wang, Zhun-ga Liu et al.

As a representative evidential clustering algorithm, evidential c-means (ECM) provides a deeper insight into the data by allowing an object to belong not only to a single class, but also to any subset of a collection of classes, which generalizes the hard, fuzzy, possibilistic, and rough partitions. However, compared with other partition-based algorithms, ECM must estimate numerous additional parameters, and thus insufficient or contaminated data will have a greater influence on its clustering performance. To solve this problem, in this study, a transfer learning-based ECM (TECM) algorithm is proposed by introducing the strategy of transfer learning into the process of evidential clustering. The TECM objective function is constructed by integrating the knowledge learned from the source domain with the data in the target domain to cluster the target data. Subsequently, an alternate optimization scheme is developed to solve the constraint objective function of the TECM algorithm. The proposed TECM algorithm is applicable to cases where the source and target domains have the same or different numbers of clusters. A series of experiments were conducted on both synthetic and real datasets, and the experimental results demonstrated the effectiveness of the proposed TECM algorithm compared to ECM and other representative multitask or transfer-clustering algorithms.

Zaidao Wen, Jiaxiang Liu, Zhunga Liu et al.

This letter presents a novel framework termed DistSTN for the task of synthetic aperture radar (SAR) automatic target recognition (ATR). In contrast to the conventional SAR ATR algorithms, DistSTN considers a more challenging practical scenario for non-cooperative targets whose aspect angles for training are incomplete and limited in a partial range while those of testing samples are unlimited. To address this issue, instead of learning the pose invariant features, DistSTN newly involves an elaborated feature disentangling model to separate the learned pose factors of a SAR target from the identity ones so that they can independently control the representation process of the target image. To disentangle the explainable pose factors, we develop a pose discrepancy spatial transformer module in DistSTN to characterize the intrinsic transformation between the factors of two different targets with an explicit geometric model. Furthermore, DistSTN develops an amortized inference scheme that enables efficient feature extraction and recognition using an encoder-decoder mechanism. Experimental results with the moving and stationary target acquisition and recognition (MSTAR) benchmark demonstrate the effectiveness of our proposed approach. Compared with the other ATR algorithms, DistSTN can achieve higher recognition accuracy.

Lianmeng Jiao, Thierry Denoeux, Zhun-ga Liu et al.

The Gaussian mixture model (GMM) provides a simple yet principled framework for clustering, with properties suitable for statistical inference. In this paper, we propose a new model-based clustering algorithm, called EGMM (evidential GMM), in the theoretical framework of belief functions to better characterize cluster-membership uncertainty. With a mass function representing the cluster membership of each object, the evidential Gaussian mixture distribution composed of the components over the powerset of the desired clusters is proposed to model the entire dataset. The parameters in EGMM are estimated by a specially designed Expectation-Maximization (EM) algorithm. A validity index allowing automatic determination of the proper number of clusters is also provided. The proposed EGMM is as simple as the classical GMM, but can generate a more informative evidential partition for the considered dataset. The synthetic and real dataset experiments show that the proposed EGMM performs better than other representative clustering algorithms. Besides, its superiority is also demonstrated by an application to multi-modal brain image segmentation.

Kuang Zhou, Arnaud Martin, Quan Pan et al.

Community detection has attracted considerable attention crossing many areas as it can be used for discovering the structure and features of complex networks. With the increasing size of social networks in real world, community detection approaches should be fast and accurate. The Label Propagation Algorithm (LPA) is known to be one of the near-linear solutions and benefits of easy implementation, thus it forms a good basis for efficient community detection methods. In this paper, we extend the update rule and propagation criterion of LPA in the framework of belief functions. A new community detection approach, called Evidential Label Propagation (ELP), is proposed as an enhanced version of conventional LPA. The node influence is first defined to guide the propagation process. The plausibility is used to determine the domain label of each node. The update order of nodes is discussed to improve the robustness of the method. ELP algorithm will converge after the domain labels of all the nodes become unchanged. The mass assignments are calculated finally as memberships of nodes. The overlapping nodes and outliers can be detected simultaneously through the proposed method. The experimental results demonstrate the effectiveness of ELP.

Kuang Zhou, Arnaud Martin, Quan Pan et al.

In this work, a new prototype-based clustering method named Evidential C-Medoids (ECMdd), which belongs to the family of medoid-based clustering for proximity data, is proposed as an extension of Fuzzy C-Medoids (FCMdd) on the theoretical framework of belief functions. In the application of FCMdd and original ECMdd, a single medoid (prototype), which is supposed to belong to the object set, is utilized to represent one class. For the sake of clarity, this kind of ECMdd using a single medoid is denoted by sECMdd. In real clustering applications, using only one pattern to capture or interpret a class may not adequately model different types of group structure and hence limits the clustering performance. In order to address this problem, a variation of ECMdd using multiple weighted medoids, denoted by wECMdd, is presented. Unlike sECMdd, in wECMdd objects in each cluster carry various weights describing their degree of representativeness for that class. This mechanism enables each class to be represented by more than one object. Experimental results in synthetic and real data sets clearly demonstrate the superiority of sECMdd and wECMdd. Moreover, the clustering results by wECMdd can provide richer information for the inner structure of the detected classes with the help of prototype weights.

Zhun-Ga Liu, Quan Pan, Jean Dezert et al.

In classification of incomplete pattern, the missing values can either play a crucial role in the class determination, or have only little influence (or eventually none) on the classification results according to the context. We propose a credal classification method for incomplete pattern with adaptive imputation of missing values based on belief function theory. At first, we try to classify the object (incomplete pattern) based only on the available attribute values. As underlying principle, we assume that the missing information is not crucial for the classification if a specific class for the object can be found using only the available information. In this case, the object is committed to this particular class. However, if the object cannot be classified without ambiguity, it means that the missing values play a main role for achieving an accurate classification. In this case, the missing values will be imputed based on the K-nearest neighbor (K-NN) and self-organizing map (SOM) techniques, and the edited pattern with the imputation is then classified. The (original or edited) pattern is respectively classified according to each training class, and the classification results represented by basic belief assignments are fused with proper combination rules for making the credal classification. The object is allowed to belong with different masses of belief to the specific classes and meta-classes (which are particular disjunctions of several single classes). The credal classification captures well the uncertainty and imprecision of classification, and reduces effectively the rate of misclassifications thanks to the introduction of meta-classes. The effectiveness of the proposed method with respect to other classical methods is demonstrated based on several experiments using artificial and real data sets.

Kuang Zhou, Arnaud Martin, Quan Pan et al.

In real clustering applications, proximity data, in which only pairwise similarities or dissimilarities are known, is more general than object data, in which each pattern is described explicitly by a list of attributes. Medoid-based clustering algorithms, which assume the prototypes of classes are objects, are of great value for partitioning relational data sets. In this paper a new prototype-based clustering method, named Evidential C-Medoids (ECMdd), which is an extension of Fuzzy C-Medoids (FCMdd) on the theoretical framework of belief functions is proposed. In ECMdd, medoids are utilized as the prototypes to represent the detected classes, including specific classes and imprecise classes. Specific classes are for the data which are distinctly far from the prototypes of other classes, while imprecise classes accept the objects that may be close to the prototypes of more than one class. This soft decision mechanism could make the clustering results more cautious and reduce the misclassification rates. Experiments in synthetic and real data sets are used to illustrate the performance of ECMdd. The results show that ECMdd could capture well the uncertainty in the internal data structure. Moreover, it is more robust to the initializations compared with FCMdd.

Kuang Zhou, Arnaud Martin, Quan Pan et al.

Median clustering is of great value for partitioning relational data. In this paper, a new prototype-based clustering method, called Median Evidential C-Means (MECM), which is an extension of median c-means and median fuzzy c-means on the theoretical framework of belief functions is proposed. The median variant relaxes the restriction of a metric space embedding for the objects but constrains the prototypes to be in the original data set. Due to these properties, MECM could be applied to graph clustering problems. A community detection scheme for social networks based on MECM is investigated and the obtained credal partitions of graphs, which are more refined than crisp and fuzzy ones, enable us to have a better understanding of the graph structures. An initial prototype-selection scheme based on evidential semi-centrality is presented to avoid local premature convergence and an evidential modularity function is defined to choose the optimal number of communities. Finally, experiments in synthetic and real data sets illustrate the performance of MECM and show its difference to other methods.