Self-supervised Learning (SSL) has become a powerful paradigm for representation learning without manual annotations. However, most existing frameworks focus on global alignment and struggle to capture the hierarchical, multi-scale lesion patterns characteristic of plant disease imagery. To address this gap, we propose PSMamba , a progressive self-supervised framework that integrates the efficient sequence modelling of Vision Mamba (VM) with a dual-student hierarchical distillation strategy. Unlike conventional single teacher-student designs, PSMamba employs a shared global teacher and two specialised students: one processes mid-scale views to capture lesion distributions and vein structures, while the other focuses on local views to capture fine-grained cues such as texture irregularities and early-stage lesions. This multi-granular supervision facilitates the joint learning of contextual and detailed representations, with consistency losses ensuring coherent cross-scale alignment. Experiments on three benchmark datasets show that PSMamba consistently outperforms representative CNN-, Transformer-, SSL-, and Mamba-based baselines, delivering superior accuracy and robustness in both domain-shifted and fine-grained scenarios.

Plant Disease Detection (PDD) is a key aspect of precision agriculture. However, existing deep learning methods often rely on extensively annotated datasets, which are time-consuming and costly to generate. Self-supervised Learning (SSL) offers a promising alternative by exploiting the abundance of unlabeled data. However, most existing SSL approaches suffer from high computational costs due to convolutional neural networks or transformer-based architectures. Additionally, they struggle to capture long-range dependencies in visual representation and rely on static loss functions that fail to align local and global features effectively. To address these challenges, we propose ConMamba, a novel SSL framework specially designed for PDD. ConMamba integrates the Vision Mamba Encoder (VME), which employs a bidirectional State Space Model (SSM) to capture long-range dependencies efficiently. Furthermore, we introduce a dual-level contrastive loss with dynamic weight adjustment to optimize local-global feature alignment. Experimental results on three benchmark datasets demonstrate that ConMamba significantly outperforms state-of-the-art methods across multiple evaluation metrics. This provides an efficient and robust solution for PDD.

Few-shot class-incremental learning (FSCIL) aims to continually learn novel data with limited samples. One of the major challenges is the catastrophic forgetting problem of old knowledge while training the model on new data. To alleviate this problem, recent state-of-the-art methods adopt a well-trained static network with fixed parameters at incremental learning stages to maintain old knowledge. These methods suffer from the poor adaptation of the old model with new knowledge. In this work, a dynamic clustering and recovering network (DyCR) is proposed to tackle the adaptation problem and effectively mitigate the forgetting phenomena on FSCIL tasks. Unlike static FSCIL methods, the proposed DyCR network is dynamic and trainable during the incremental learning stages, which makes the network capable of learning new features and better adapting to novel data. To address the forgetting problem and improve the model performance, a novel orthogonal decomposition mechanism is developed to split the feature embeddings into context and category information. The context part is preserved and utilized to recover old class features in future incremental learning stages, which can mitigate the forgetting problem with a much smaller size of data than saving the raw exemplars. The category part is used to optimize the feature embedding space by moving different classes of samples far apart and squeezing the sample distances within the same classes during the training stage. Experiments show that the DyCR network outperforms existing methods on four benchmark datasets. The code is available at: https://github.com/zichengpan/DyCR.

Few-shot class-incremental learning was introduced to solve the model adaptation problem for new incremental classes with only a few examples while still remaining effective for old data. Although recent state-of-the-art methods make some progress in improving system robustness on common datasets, they fail to work on fine-grained datasets where inter-class differences are small. The problem is mainly caused by: (1) the overlapping of new data and old data in the feature space during incremental learning, which means old samples can be falsely classified as newly introduced classes and induce catastrophic forgetting phenomena; (2) lacking discriminative feature learning ability to identify fine-grained objects. In this paper, a novel Pseudo-set Frequency Refinement (PFR) architecture is proposed to tackle these problems. We design a pseudo-set training strategy to mimic the incremental learning scenarios so that the model can better adapt to novel data in future incremental sessions. Furthermore, separate adaptation tasks are developed by utilizing frequency-based information to refine the original features and address the above challenging problems. More specifically, the high and low-frequency components of the images are employed to enrich the discriminative feature analysis ability and incremental learning ability of the model respectively. The refined features are used to perform inter-class and inter-set analyses. Extensive experiments show that the proposed method consistently outperforms the state-of-the-art methods on four fine-grained datasets.

Ultra-Fine-Grained Visual Categorization (ultra-FGVC) has become a popular problem due to its great real-world potential for classifying the same or closely related species with very similar layouts. However, there present many challenges for the existing ultra-FGVC methods, firstly there are always not enough samples in the existing ultraFGVC datasets based on which the models can easily get overfitting. Secondly, in practice, we are likely to find new species that we have not seen before and need to add them to existing models, which is known as incremental learning. The existing methods solve these problems by Few-Shot Class Incremental Learning (FSCIL), but the main challenge of the FSCIL models on ultra-FGVC tasks lies in their inferior discrimination detection ability since they usually use low-capacity networks to extract features, which leads to insufficient discriminative details extraction from ultrafine-grained images. In this paper, a self-supervised feature enhancement for the few-shot incremental learning network (SSFE-Net) is proposed to solve this problem. Specifically, a self-supervised learning (SSL) and knowledge distillation (KD) framework is developed to enhance the feature extraction of the low-capacity backbone network for ultra-FGVC few-shot class incremental learning tasks. Besides, we for the first time create a series of benchmarks for FSCIL tasks on two public ultra-FGVC datasets and three normal finegrained datasets, which will facilitate the development of the Ultra-FGVC community. Extensive experimental results on public ultra-FGVC datasets and other state-of-the-art benchmarks consistently demonstrate the effectiveness of the proposed method.

Two-dimensional singular value decomposition (2DSVD) is an important dimensionality reduction algorithm which has inherent advantage in preserving the structure of 2D images. However, 2DSVD algorithm is based on the squared error loss, which may exaggerate the projection errors with the presence of outliers. To solve this problem, we propose a generalized kernel risk sensitive loss for measuring the projection error in 2DSVD, which automatically eliminates the outlier information during optimization. Since the proposed objective function is non-convex, a majorization-minimization algorithm is developed to efficiently solve it. Our method is rotational invariant and has intrinsic advantages in processing non-centered data. Experimental results on public databases demonstrate that the performance of the proposed method significantly outperforms several benchmark methods on different applications.

Traditional tensor decomposition methods, e.g., two dimensional principal component analysis and two dimensional singular value decomposition, that minimize mean square errors, are sensitive to outliers. To overcome this problem, in this paper we propose a new robust tensor decomposition method using generalized correntropy criterion (Corr-Tensor). A Lagrange multiplier method is used to effectively optimize the generalized correntropy objective function in an iterative manner. The Corr-Tensor can effectively improve the robustness of tensor decomposition with the existence of outliers without introducing any extra computational cost. Experimental results demonstrated that the proposed method significantly reduces the reconstruction error on face reconstruction and improves the accuracies on handwritten digit recognition and facial image clustering.

Partial occlusions in face images pose a great problem for most face recognition algorithms due to the fact that most of these algorithms mainly focus on solving a second order loss function, e.g., mean square error (MSE), which will magnify the effect from occlusion parts. In this paper, we proposed a kernel non-second order loss function for sparse representation (KNS-SR) to recognize or restore partially occluded facial images, which both take the advantages of the correntropy and the non-second order statistics measurement. The resulted framework is more accurate than the MSE-based ones in locating and eliminating outliers information. Experimental results from image reconstruction and recognition tasks on publicly available databases show that the proposed method achieves better performances compared with existing methods.

Traditional matrix-based dimensional reduction methods, e.g., two-dimensional principal component analysis (2DPCA) and two-dimensional singular value decomposition (2DSVD), minimize mean square errors (MSE), which is sensitive to outliers. To overcome this problem, in this paper we propose a new robust 2DSVD method based on the kernel mean p power error loss (KMPE-2DSVD). Different from the MSE and the correntropy based ones which are second order statistics based measurements, the KMPE-2DSVD is based on the non-second order statistics in the kernel space, and thus is more flexible in controlling the representation error. Experimental results show that the proposed method significantly improves the accuracy of facial image clustering.

The orthogonal matching pursuit (OMP) is an important sparse approximation algorithm to recover sparse signals from compressed measurements. However, most MP algorithms are based on the mean square error(MSE) to minimize the recovery error, which is suboptimal when there are outliers. In this paper, we present a new robust OMP algorithm based on kernel non-second order statistics (KNS-OMP), which not only takes advantages of the outlier resistance ability of correntropy but also further extends the second order statistics based correntropy to a non-second order similarity measurement to improve its robustness. The resulted framework is more accurate than the second order ones in reducing the effect of outliers. Experimental results on synthetic and real data show that the proposed method achieves better performances compared with existing methods.

Traditional tensor decomposition methods, e.g., two dimensional principle component analysis (2DPCA) and two dimensional singular value decomposition (2DSVD), minimize mean square errors (MSE) and are sensitive to outliers. In this paper, we propose a new robust tensor factorization method using maximum correntropy criterion (MCC) to improve the robustness of traditional tensor decomposition methods. A half-quadratic optimization algorithm is adopted to effectively optimize the correntropy objective function in an iterative manner. It can effectively improve the robustness of a tensor decomposition method to outliers without introducing any extra computational cost. Experimental results demonstrated that the proposed method significantly reduces the reconstruction error on face reconstruction and improves the accuracy rate on handwritten digit recognition.

Shape fitting, including straight line and ellipse fitting, plays an important role in the (cylinder-) shape control system of silicon single crystal growth, because the straight lines and ellipse in the crystal image contain the important horizontal circle center and diameter information. This information can be used as control variables so that the grown crystal approximates to a perfect cylinder, and thus can be used as high-quality source materials. In this paper, we develop new straight line and ellipse fitting algorithms. The key points are as follows. We formulate the two-dimensional (2-D) binary image into a single-snapshot array signal of a virtual sensor array, and casts the angle estimation problem of straight lines into the direction finding one of virtual incoming sources. Based on the virtual array manifold and potential incoming angles, the relevant over-complete dictionary is constructed, and thus a sparse regression problem is formed. To solve such a regression problem, we introduce the weight vector sparsity term into the conventional linear least-squares support vector regression framework to estimate the angles of these straight lines. Based on the estimated angles and potential offsets, another over-complete dictionary is constructed, and thus the image can be looked upon as the sparse representation of these dictionary atoms. Since the constructed dictionary is of the same size as the image, we use the compressed sensing theory to reduce the relevant dimensionality and then apply the aforementioned sparse regression method to obtain the relevant offsets of these straight lines. We derive a new second-order polynomial of ellipse equation to obtain the ellipse parameters to avoid the trival solution from the conventional polynomial model. Some simulation and experimental examples are given to illustrate the effectiveness of the proposed algorithms.

This paper develops a distributed dictionary learning algorithm for sparse representation of the data distributed across nodes of sensor networks, where the sensitive or private data are stored or there is no fusion center or there exists a big data application. The main contributions of this paper are: 1) we decouple the combined dictionary atom update and nonzero coefficient revision procedure into two-stage operations to facilitate distributed computations, first updating the dictionary atom in terms of the eigenvalue decomposition of the sum of the residual (correlation) matrices across the nodes then implementing a local projection operation to obtain the related representation coefficients for each node; 2) we cast the aforementioned atom update problem as a set of decentralized optimization subproblems with consensus constraints. Then, we simplify the multiplier update for the symmetry undirected graphs in sensor networks and minimize the separable subproblems to attain the consistent estimates iteratively; and 3) dictionary atoms are typically constrained to be of unit norm in order to avoid the scaling ambiguity. We efficiently solve the resultant hidden convex subproblems by determining the optimal Lagrange multiplier. Some experiments are given to show that the proposed algorithm is an alternative distributed dictionary learning approach, and is suitable for the sensor network environment.

Ellipse fitting is widely applied in the fields of computer vision and automatic industry control, in which the procedure of ellipse fitting often follows the preprocessing step of edge detection in the original image. Therefore, the ellipse fitting method also depends on the accuracy of edge detection besides their own performance, especially due to the introduced outliers and edge point errors from edge detection which will cause severe performance degradation. In this paper, we develop a robust ellipse fitting method to alleviate the influence of outliers. The proposed algorithm solves ellipse parameters by linearly combining a subset of ("more accurate") data points (formed from edge points) rather than all data points (which contain possible outliers). In addition, considering that squaring the fitting residuals can magnify the contributions of these extreme data points, our algorithm replaces it with the absolute residuals to reduce this influence. Moreover, the norm of data point errors is bounded, and the worst case performance optimization is formed to be robust against data point errors. The resulting mixed l1-l2 optimization problem is further derived as a second-order cone programming one and solved by the computationally efficient interior-point methods. Note that the fitting approach developed in this paper specifically deals with the overdetermined system, whereas the current sparse representation theory is only applied to underdetermined systems. Therefore, the proposed algorithm can be looked upon as an extended application and development of the sparse representation theory. Some simulated and experimental examples are presented to illustrate the effectiveness of the proposed ellipse fitting approach.