Transformer has been extensively explored for hyperspectral image (HSI) classification. However, transformer poses challenges in terms of speed and memory usage because of its quadratic computational complexity. Recently, the Mamba model has emerged as a promising approach, which has strong long-distance modeling capabilities while maintaining a linear computational complexity. However, representing the HSI is challenging for the Mamba due to the requirement for an integrated spatial and spectral understanding. To remedy these drawbacks, we propose a novel HSI classification model based on a Mamba model, named MambaHSI, which can simultaneously model long-range interaction of the whole image and integrate spatial and spectral information in an adaptive manner. Specifically, we design a spatial Mamba block (SpaMB) to model the long-range interaction of the whole image at the pixel-level. Then, we propose a spectral Mamba block (SpeMB) to split the spectral vector into multiple groups, mine the relations across different spectral groups, and extract spectral features. Finally, we propose a spatial-spectral fusion module (SSFM) to adaptively integrate spatial and spectral features of a HSI. To our best knowledge, this is the first image-level HSI classification model based on the Mamba. We conduct extensive experiments on four diverse HSI datasets. The results demonstrate the effectiveness and superiority of the proposed model for HSI classification. This reveals the great potential of Mamba to be the next-generation backbone for HSI models. Codes are available at https://github.com/li-yapeng/MambaHSI .
Multi-Dimensional time series classification and prediction has been widely used in many fields, such as disease prevention, fault diagnosis and action recognition. However, the traditional method needs manual intervention and inference, and cannot realize the figurative expression of multi-Dimensional data, which lead to inadequate information mining. Inspired by the strong power of deep learning technology in image processing, we propose a unified time-series image fusion framework to transform multi-modal data into 2D-image, and then realize automatic feature extraction and classification based on a lightweight convolutional neural network. We present two basic image coding methods, Gray image coding, RGB image coding, and their step coding methods. Considering the universality of different application fields, we extended the coding method and propose two types transform coding, Transform-RGB coding and RGB-Transform coding, to improve the multi-domain representation ability. By applying to three typical scenes of Parkinson's disease diagnosis, bearing fault detection and gymnastics action recognition, we obtained the highest classification accuracy of 100%, 92.86% and 99.70% respectively, which were all higher than the classical processing methods. It proves the strong classification ability and universality of our coding framework to different multi-dimensional scenes. We expect that this method can be used and perform well in other scenarios, and be potential to facilitate the progress of related technology.
Medical image classification is one of the most critical problems in the image recognition area. One of the major challenges in this field is the scarcity of labelled training data. Additionally, there is often class imbalance in datasets as some cases are very rare to happen. As a result, accuracy in classification task is normally low. Deep Learning models, in particular, show promising results on image segmentation and classification problems, but they require very large datasets for training. Therefore, there is a need to generate more of synthetic samples from the same distribution. Previous work has shown that feature generation is more efficient and leads to better performance than corresponding image generation. We apply this idea in the Medical Imaging domain. We use transfer learning to train a segmentation model for the small dataset for which gold-standard class annotations are available. We extracted the learnt features and use them to generate synthetic features conditioned on class labels, using Auxiliary Classifier GAN (ACGAN). We test the quality of the generated features in a downstream classification task for brain tumors according to their severity level. Experimental results show a promising result regarding the validity of these generated features and their overall contribution to balancing the data and improving the classification class-wise accuracy.
Image segmentation is an important problem in many safety-critical applications. Recent studies show that modern image segmentation models are vulnerable to adversarial perturbations, while existing attack methods mainly follow the idea of attacking image classification models. We argue that image segmentation and classification have inherent differences, and design an attack framework specially for image segmentation models. Our attack framework is inspired by certified radius, which was originally used by defenders to defend against adversarial perturbations to classification models. We are the first, from the attacker perspective, to leverage the properties of certified radius and propose a certified radius guided attack framework against image segmentation models. Specifically, we first adapt randomized smoothing, the state-of-the-art certification method for classification models, to derive the pixel's certified radius. We then focus more on disrupting pixels with relatively smaller certified radii and design a pixel-wise certified radius guided loss, when plugged into any existing white-box attack, yields our certified radius-guided white-box attack. Next, we propose the first black-box attack to image segmentation models via bandit. We design a novel gradient estimator, based on bandit feedback, which is query-efficient and provably unbiased and stable. We use this gradient estimator to design a projected bandit gradient descent (PBGD) attack, as well as a certified radius-guided PBGD (CR-PBGD) attack. We prove our PBGD and CR-PBGD attacks can achieve asymptotically optimal attack performance with an optimal rate. We evaluate our certified-radius guided white-box and black-box attacks on multiple modern image segmentation models and datasets. Our results validate the effectiveness of our certified radius-guided attack framework.
The thinnest possible camera is achieved by removing all optics, leaving only the image sensor. We train deep neural networks to perform multi-class detection and binary classification (with accuracy of 92%) on optics-free images without the need for anthropocentric image reconstructions. Inferencing from optics-free images has the potential for enhanced privacy and power efficiency.
Scene text instances found in natural images carry explicit semantic information that can provide important cues to solve a wide array of computer vision problems. In this paper, we focus on leveraging multi-modal content in the form of visual and textual cues to tackle the task of fine-grained image classification and retrieval. First, we obtain the text instances from images by employing a text reading system. Then, we combine textual features with salient image regions to exploit the complementary information carried by the two sources. Specifically, we employ a Graph Convolutional Network to perform multi-modal reasoning and obtain relationship-enhanced features by learning a common semantic space between salient objects and text found in an image. By obtaining an enhanced set of visual and textual features, the proposed model greatly outperforms the previous state-of-the-art in two different tasks, fine-grained classification and image retrieval in the Con-Text and Drink Bottle datasets.
In the field of vision-language contrastive learning, models such as CLIP capitalize on matched image-caption pairs as positive examples and leverage within-batch non-matching pairs as negatives. This approach has led to remarkable outcomes in zero-shot image classification, cross-modal retrieval, and linear evaluation tasks. We show that the zero-shot classification and retrieval capabilities of CLIP-like models can be improved significantly through the introduction of semantically composite examples during pretraining. Inspired by CutMix in vision categorization, we create semantically composite image-caption pairs by merging elements from two distinct instances in the dataset via a novel procedure. Our method fuses the captions and blends 50% of each image to form a new composite sample. This simple technique (termed CLIP-C for CLIP Compositions), devoid of any additional computational overhead or increase in model parameters, significantly improves zero-shot image classification and cross-modal retrieval. The benefits of CLIP-C are particularly pronounced in settings with relatively limited pretraining data.
Dimensionality reduction is a crucial preprocessing for hyperspectral data analysis - finding an appropriate subspace is often required for subsequent image classification. In recent work, we proposed supervised angular information based dimensionality reduction methods to find effective subspaces. Since unlabeled data are often more readily available compared to labeled data, we propose an unsupervised projection that finds a lower dimensional subspace where local angular information is preserved. To exploit spatial information from the hyperspectral images, we further extend our unsupervised projection to incorporate spatial contextual information around each pixel in the image. Additionally, we also propose a sparse representation based classifier which is optimized to exploit spatial information during classification - we hence assert that our proposed projection is particularly suitable for classifiers where local similarity and spatial context are both important. Experimental results with two real-world hyperspectral datasets demonstrate that our proposed methods provide a robust classification performance.
In this work we developed a novel and fast traffic sign recognition system, a very important part for advanced driver assistance system and for autonomous driving. Traffic signs play a very vital role in safe driving and avoiding accident. We have used image processing and topic discovery model pLSA to tackle this challenging multiclass classification problem. Our algorithm is consist of two parts, shape classification and sign classification for improved accuracy. For processing and representation of image we have used bag of features model with SIFT local descriptor. Where a visual vocabulary of size 300 words are formed using k-means codebook formation algorithm. We exploited the concept that every image is a collection of visual topics and images having same topics will belong to same category. Our algorithm is tested on German traffic sign recognition benchmark (GTSRB) and gives very promising result near to existing state of the art techniques.
In this paper the Modified Fractal Signature method is applied to real Synthetic Aperture Radar images provided to our research group by SET 163 Working Group on SAR radar techniques. This method uses the blanket technique to provide useful information for SAR image classification. It is based on the calculation of the volume of a blanket, corresponding to the image to be classified, and then on the calculation of the corresponding Fractal Area curve and Fractal Dimension curve of the image. The main idea concerning this proposed technique is the fact that different terrain types encountered in SAR images yield different values of Fractal Area curves and Fractal Dimension curves, upon which classification of different types of terrain is possible. As a result, a classification technique for five different terrain types, i.e. urban, suburban, rural, mountain and sea, is presented in this paper.
In spite of achieving revolutionary successes in machine learning, deep convolutional neural networks have been recently found to be vulnerable to adversarial attacks and difficult to generalize to novel test images with reasonably large geometric transformations. Inspired by a recent neuroscience discovery revealing that primate brain employs disentangled shape and appearance representations for object recognition, we propose a general disentangled deep autoencoding regularization framework that can be easily applied to any deep embedding based classification model for improving the robustness of deep neural networks. Our framework effectively learns disentangled appearance code and geometric code for robust image classification, which is the first disentangling based method defending against adversarial attacks and complementary to standard defense methods. Extensive experiments on several benchmark datasets show that, our proposed regularization framework leveraging disentangled embedding significantly outperforms traditional unregularized convolutional neural networks for image classification on robustness against adversarial attacks and generalization to novel test data.
Neural networks have been notorious for being computationally expensive. This is mainly because neural networks are often over-parametrized and most likely have redundant nodes or layers as they are getting deeper and wider. Their demand for hardware resources prohibits their extensive use in embedded devices and puts restrictions on tasks like real-time image classification or object detection. In this work, we propose a network-agnostic model compression method infused with a novel dynamical clustering approach to reduce the computational cost and memory footprint of deep neural networks. We evaluated our new compression method on five different state-of-the-art image classification and object detection networks. In classification networks, we pruned about 95% of network parameters. In advanced detection networks such as YOLOv3, our proposed compression method managed to reduce the model parameters up to 59.70% which yielded 110X less memory without sacrificing much in accuracy.
Deep learning based methods have seen a massive rise in popularity for hyperspectral image classification over the past few years. However, the success of deep learning is attributed greatly to numerous labeled samples. It is still very challenging to use only a few labeled samples to train deep learning models to reach a high classification accuracy. An active deep-learning framework trained by an end-to-end manner is, therefore, proposed by this paper in order to minimize the hyperspectral image classification costs. First, a deep densely connected convolutional network is considered for hyperspectral image classification. Different from the traditional active learning methods, an additional network is added to the designed deep densely connected convolutional network to predict the loss of input samples. Then, the additional network could be used to suggest unlabeled samples that the deep densely connected convolutional network is more likely to produce a wrong label. Note that the additional network uses the intermediate features of the deep densely connected convolutional network as input. Therefore, the proposed method is an end-to-end framework. Subsequently, a few of the selected samples are labelled manually and added to the training samples. The deep densely connected convolutional network is therefore trained using the new training set. Finally, the steps above are repeated to train the whole framework iteratively. Extensive experiments illustrates that the method proposed could reach a high accuracy in classification after selecting just a few samples.
Background and Purpose: Convolutional neural network is widely used for image recognition in the medical area at nowadays. However, overall accuracy in predicting lung tumor is low and the processing time is high as the error occurred while reconstructing the CT image. The aim of this work is to increase the overall prediction accuracy along with reducing processing time by using multispace image in pooling layer of convolution neural network. Methodology: The proposed method has the autoencoder system to improve the overall accuracy, and to predict lung cancer by using multispace image in pooling layer of convolution neural network and Adam Algorithm for optimization. First, the CT images were pre-processed by feeding image to the convolution filter and down sampled by using max pooling. Then, features are extracted using the autoencoder model based on convolutional neural network and multispace image reconstruction technique is used to reduce error while reconstructing the image which then results improved accuracy to predict lung nodule. Finally, the reconstructed images are taken as input for SoftMax classifier to classify the CT images. Results: The state-of-art and proposed solutions were processed in Python Tensor Flow and It provides significant increase in accuracy in classification of lung cancer to 99.5 from 98.9 and decrease in processing time from 10 frames/second to 12 seconds/second. Conclusion: The proposed solution provides high classification accuracy along with less processing time compared to the state of art. For future research, large dataset can be implemented, and low pixel image can be processed to evaluate the classification
In this work, we compare the performance of six state-of-the-art deep neural networks in classification tasks when using only image features, to when these are combined with patient metadata. We utilise transfer learning from networks pretrained on ImageNet to extract image features from the ISIC HAM10000 dataset prior to classification. Using several classification performance metrics, we evaluate the effects of including metadata with the image features. Furthermore, we repeat our experiments with data augmentation. Our results show an overall enhancement in performance of each network as assessed by all metrics, only noting degradation in a vgg16 architecture. Our results indicate that this performance enhancement may be a general property of deep networks and should be explored in other areas. Moreover, these improvements come at a negligible additional cost in computation time, and therefore are a practical method for other applications.
We propose a new way to explain and to visualize neural network classification through a decomposition-based explainable AI (DXAI). Instead of providing an explanation heatmap, our method yields a decomposition of the image into class-agnostic and class-distinct parts, with respect to the data and chosen classifier. Following a fundamental signal processing paradigm of analysis and synthesis, the original image is the sum of the decomposed parts. We thus obtain a radically different way of explaining classification. The class-agnostic part ideally is composed of all image features which do not posses class information, where the class-distinct part is its complementary. This new visualization can be more helpful and informative in certain scenarios, especially when the attributes are dense, global and additive in nature, for instance, when colors or textures are essential for class distinction. Code is available at https://github.com/dxai2024/dxai.
This paper addresses the land cover classification task for remote sensing images by deep self-taught learning. Our self-taught learning approach learns suitable feature representations of the input data using sparse representation and undercomplete dictionary learning. We propose a deep learning framework which extracts representations in multiple layers and use the output of the deepest layer as input to a classification algorithm. We evaluate our approach using a multispectral Landsat 5 TM image of a study area in the North of Novo Progresso (South America) and the Zurich Summer Data Set provided by the University of Zurich. Experiments indicate that features learned by a deep self-taught learning framework can be used for classification and improve the results compared to classification results using the original feature representation.
In this study, we propose a fast and accurate method to automatically localize anatomical landmarks in medical images. We employ a global-to-local localization approach using fully convolutional neural networks (FCNNs). First, a global FCNN localizes multiple landmarks through the analysis of image patches, performing regression and classification simultaneously. In regression, displacement vectors pointing from the center of image patches towards landmark locations are determined. In classification, presence of landmarks of interest in the patch is established. Global landmark locations are obtained by averaging the predicted displacement vectors, where the contribution of each displacement vector is weighted by the posterior classification probability of the patch that it is pointing from. Subsequently, for each landmark localized with global localization, local analysis is performed. Specialized FCNNs refine the global landmark locations by analyzing local sub-images in a similar manner, i.e. by performing regression and classification simultaneously and combining the results. Evaluation was performed through localization of 8 anatomical landmarks in CCTA scans, 2 landmarks in olfactory MR scans, and 19 landmarks in cephalometric X-rays. We demonstrate that the method performs similarly to a second observer and is able to localize landmarks in a diverse set of medical images, differing in image modality, image dimensionality, and anatomical coverage.
Open-vocabulary models are a promising new paradigm for image classification. Unlike traditional classification models, open-vocabulary models classify among any arbitrary set of categories specified with natural language during inference. This natural language, called "prompts", typically consists of a set of hand-written templates (e.g., "a photo of a {}") which are completed with each of the category names. This work introduces a simple method to generate higher accuracy prompts, without relying on any explicit knowledge of the task domain and with far fewer hand-constructed sentences. To achieve this, we combine open-vocabulary models with large language models (LLMs) to create Customized Prompts via Language models (CuPL, pronounced "couple"). In particular, we leverage the knowledge contained in LLMs in order to generate many descriptive sentences that contain important discriminating characteristics of the image categories. This allows the model to place a greater importance on these regions in the image when making predictions. We find that this straightforward and general approach improves accuracy on a range of zero-shot image classification benchmarks, including over one percentage point gain on ImageNet. Finally, this simple baseline requires no additional training and remains completely zero-shot. Code available at https://github.com/sarahpratt/CuPL.
Due to the swift growth of patent applications each year, information and multimedia retrieval approaches that facilitate patent exploration and retrieval are of utmost importance. Different types of visualizations (e.g., graphs, technical drawings) and perspectives (e.g., side view, perspective) are used to visualize details of innovations in patents. The classification of these images enables a more efficient search and allows for further analysis. So far, datasets for image type classification miss some important visualization types for patents. Furthermore, related work does not make use of recent deep learning approaches including transformers. In this paper, we adopt state-of-the-art deep learning methods for the classification of visualization types and perspectives in patent images. We extend the CLEF-IP dataset for image type classification in patents to ten classes and provide manual ground truth annotations. In addition, we derive a set of hierarchical classes from a dataset that provides weakly-labeled data for image perspectives. Experimental results have demonstrated the feasibility of the proposed approaches. Source code, models, and dataset will be made publicly available.