Search

Seagrass ecosystems are pivotal in marine environments, serving as crucial habitats for diverse marine species and contributing significantly to carbon sequestration. Accurate classification of seagrass species from underwater images is imperative for monitoring and preserving these ecosystems. This paper introduces Unsupervised Curriculum Learning (UCL) to seagrass classification using the DeepSeagrass dataset. UCL progressively learns from simpler to more complex examples, enhancing the model's ability to discern seagrass features in a curriculum-driven manner. Experiments employing state-of-the-art deep learning architectures, convolutional neural networks (CNNs), show that UCL achieved overall 90.12 % precision and 89 % recall, which significantly improves classification accuracy and robustness, outperforming some traditional supervised learning approaches like SimCLR, and unsupervised approaches like Zero-shot CLIP. The methodology of UCL involves four main steps: high-dimensional feature extraction, pseudo-label generation through clustering, reliable sample selection, and fine-tuning the model. The iterative UCL framework refines CNN's learning of underwater images, demonstrating superior accuracy, generalization, and adaptability to unseen seagrass and background samples of undersea images. The findings presented in this paper contribute to the advancement of seagrass classification techniques, providing valuable insights into the conservation and management of marine ecosystems. The code and dataset are made publicly available and can be assessed here: https://github.com/nabid69/Unsupervised-Curriculum-Learning—UCL.

Localisation of clinically relevant events within Electroencephalogram (EEG) recordings can be useful for explaining the decisions made by automated EEG screening and decision support systems. The majority of existing deep learning based approaches that have been proposed in recent literature only classify EEG records as normal or pathological without providing any justification for their decisions and thus are not very transparent. In clinical practice it is often observed that a significant proportion of EEG recordings does not contain any abnormal (or pathological) events; even in cases classified as pathological. If deployed in practice such a setup would not be very useful since it would require neurologists to invest additional time, manually searching for events within an EEG recording before accepting or rejecting the decision proposed by the automated system. This work presents open-source software that can automatically localise and classify abnormalities both across time and EEG channels. Our work can thus be used to reveal the reasons behind an EEG recording being classified as normal or pathological/abnormal. Training an automated event localisation system requires a dataset containing fine-grained labels pointing out precise locations of events. To facilitate further development we are also releasing the dataset and annotations used in this work for use by the research community. This dataset contains 1,075 EEG recordings with precise temporal and channel locations of two broad categories of abnormal events: (i) Epileptiform discharges and (ii) Non-epileptiform abnormalities. Our localisation system is based on features derived from wavelet transforms. For event classification we investigated the performance of both classic machine learning algorithms (support vector machines, decision trees, random forest classifier) and deep convolutional neural networks (VGG16, GoogLeNet and EfficientNet). Our results indicate that deep convolutional neural networks outperform classic machine learning algorithms in terms of average values of precision, recall, F1-score and accuracy.

Among the numerous techniques followed to learn a linear classifier through the discriminative dictionary and sparse representations learning of signals, the techniques to learn a nonparametric Bayesian classifier jointly and discriminately with the dictionary and the corresponding sparse representations have drawn considerable attention from researchers. These techniques jointly learn two sets of sparse representations, one for the training samples over the dictionary and the other for the corresponding labels over the dictionary classifier. At the prediction stage, the representations of the test samples computed over the learned dictionary do not truly represent the corresponding labels, exposing weakness in the joint learning claim of these techniques. We mitigate this problem and strengthen the joint by learning a set of weights over the dictionary to represent the training data and further optimizing the same weights over the dictionary classifier to represent the labels of the corresponding classes of the training data. Now, at the prediction stage, the representation weights of the test samples computed over the learned dictionary also represent the labels of the corresponding classes of the test samples, resulting in the accurate reconstruction of the labels of the classes by the learned dictionary classifier. Overall, a reduction in the size of the Bayesian model’s parameters also improves training time. We analytically and nonparametrically derived the posterior conditional probabilities of the model from the overall joint probability of the model using Bayes’ theorem. We used the Gibbs sampler to solve the joint probability of the model using the derived conditional probabilities, which also supports our claim of efficient optimization of the coupled/joint dictionaries and the sparse representation parameters. We demonstrated the effectiveness of our approach through experiments on the standard datasets, i.e., the Extended YaleB and AR face databases for face recognition, Caltech-101 and Fifteen Scene Category databases for categorization, and UCF sports action database for action recognition. We compared the results with the state-of-the-art methods in the area. The classification accuracies, i.e., 93.25%, 89.27%, 94.81%, 98.10%, and 95.00%, of our approach on the datasets have increases of 0.5 to 2% on average. The overall average error margin of the confidence intervals in our approach is 0.24 compared with the second-best approach, JBDC, for which it is 0.34. The AUC–ROC scores of our approach are 0.98 and 0.992, which are better than those of others, i.e., 0.960 and 0.98, respectively. Our approach is also computationally efficient.

Deep neural networks can be fooled by small imperceptible perturbations called adversarial examples. Although these examples are carefully crafted, they involve two major concerns. In some cases, adversarial examples generated are much larger than minimal adversarial perturbations while in others the attack method involves an extensive number of iterations making it infeasible. Moreover, the sparse attacks are either too complex or are not sparse enough to achieve imperceptibility. Therefore, attacks designed should be fast and minimum in terms of $\ell _{2}$ -norm. In this research, we used a dictionary learning technique to generate sparse adversarial examples based on feature maps of target images. We present two novel algorithms to tune the dictionary learning process and feature map selection. The results on MNIST and Imagenet show our attack is better or competitive with the state-of-the-art methods. We also compared our method with sparse attacks recently introduced in literature. As a result, we have achieved comparable attack success rate when compared to the state-of-the-art with smaller $\ell _{2}$ -norm. We also tested the efficacy of our attack in the presence of defense mechanisms and none of the defenses were able to combat the effect of our proposed attack

Wheat stripe rust disease (WRD) is extremely detrimental to wheat crop health, and it severely affects the crop yield, increasing the risk of food insecurity. Manual inspection by trained personnel is carried out to inspect the disease spread and extent of damage to wheat fields. However, this is quite inefficient, time-consuming, and laborious, owing to the large area of wheat plantations. Artificial intelligence (AI) and deep learning (DL) offer efficient and accurate solutions to such real-world problems. By analyzing large amounts of data, AI algorithms can identify patterns that are difficult for humans to detect, enabling early disease detection and prevention. However, deep learning models are data-driven, and scarcity of data related to specific crop diseases is one major hindrance in developing models. To overcome this limitation, in this work, we introduce an annotated real-world semantic segmentation dataset named the NUST Wheat Rust Disease (NWRD) dataset. Multileaf images from wheat fields under various illumination conditions with complex backgrounds were collected, preprocessed, and manually annotated to construct a segmentation dataset specific to wheat stripe rust disease. Classification of WRD into different types and categories is a task that has been solved in the literature; however, semantic segmentation of wheat crops to identify the specific areas of plants and leaves affected by the disease remains a challenge. For this reason, in this work, we target semantic segmentation of WRD to estimate the extent of disease spread in wheat fields. Sections of fields where the disease is prevalent need to be segmented to ensure that the sick plants are quarantined and remedial actions are taken. This will consequently limit the use of harmful fungicides only on the targeted disease area instead of the majority of wheat fields, promoting environmentally friendly and sustainable farming solutions. Owing to the complexity of the proposed NWRD segmentation dataset, in our experiments, promising results were obtained using the UNet semantic segmentation model and the proposed adaptive patching with feedback (APF) technique, which produced a precision of 0.506, recall of 0.624, and F1 score of 0.557 for the rust class.

With the advent of smart grid, load forecasting is emerging as an essential technology to implement optimal planning and control of grid assets. Ergo in recent years, a significant thrust can be witnessed for the research towards the improvement of the prediction of the energy demand. However thus far there has not been any one technique in the literature that is shown to give best forecasts for a variety of sites; almost all the papers published on load forecasting, report their best results on just one of the dataset. This problem accentuates further when the training data does not have enough data points to learn patterns over all the seasons. Hence to devise a load forecasting technique that can yield the best estimates on diverse datasets, especially when the training data is limited, is a big challenge, which is addressed in this paper. The paper presents a novel combination of deep learning with feature engineering for short-term load forecasting. The proposed architecture, named as Deep Derived Feature Fusion (DeepDeFF), is based on the sequential model in conjunction with the hand-crafted derived features in order to aid the model for better learning and predictions. The raw data and the hand-crafted features are trained at separate levels, then their respective outputs are combined to make the final prediction. The efficacy and robustness of the proposed methodology is evaluated on diverse datasets from five countries with completely different patterns. The extensive experiments and results demonstrate that the proposed technique is superior to the existing state of the art.

The understanding and analyzing of available content on Social media Platforms such as Twitter and Facebook, through various topic modeling methods is not supervised. However, despite several existing conventional techniques, they have had limited success when applied directly for filtering and quick comprehension of short-text contents due to text sparseness and noise. Thus, it always has been challenging to discover reliable latent topics from online discussion texts that prevail with low words co-occurrence and availability of large size social media benchmark datasets, even for resource-rich languages. The existing literature lacks such work for Urdu text to unveil niche topics even with conventional topic models, mainly due to the lack of benchmark datasets, limited availability of pre-processing tools/ algorithms, and time and compute limitations on large-sized datasets. This work presents experiments with multiple approaches of topic modeling like Latent Semantic Analysis (LSA), Probabilistic Latent Semantic Analysis (PLSA), Latent Dirichlet Allocation (LDA), and Non-negative Matrix Factorization (NMF) on 0.8 million Urdu tweets. These tweets are collected through Twitter API by giving various hashtags as a query to avoid dominance of single topic in the dataset. In addition, we have pre-processed the text of the tweets, prepared the three variants of the collected dataset, and extracted multiple features to represent documents on different n-grams. Furthermore, all these techniques are compared and evaluated on the dataset variants, using both qualitative and quantitative measures. We have also demonstrated the results of these approaches through visualization methods, graphs depicting tweets size per topic, word clouds, and hashtags analysis, giving insights about algorithms performances on finalized topics. Observed results reveal that NMF outperformed the techniques with TF-IDF feature vectors in Urdu tweets text, while LDA performed best with merging short-text strategy into long pseudo documents.

Electroencephalogram (EEG) is widely used for the diagnosis of neurological conditions like epilepsy, neurodegenerative illnesses and sleep related disorders. Proper interpretation of EEG recordings requires the expertise of trained neurologists, a resource which is scarce in the developing world. Neurologists spend a significant portion of their time sifting through EEG recordings looking for abnormalities. Most recordings turn out to be completely normal, owing to the low yield of EEG tests. To minimize such wastage of time and effort, automatic algorithms could be used to provide pre-diagnostic screening to separate normal from abnormal EEG. Data driven machine learning offers a way forward however, design and verification of modern machine learning algorithms require properly curated labeled datasets. To avoid bias, deep learning based methods must be trained on large datasets from diverse sources. This work presents a new open-source dataset, named the NMT Scalp EEG Dataset, consisting of 2,417 recordings from unique participants spanning almost 625 h. Each recording is labeled as normal or abnormal by a team of qualified neurologists. Demographic information such as gender and age of the patient are also included. Our dataset focuses on the South Asian population. Several existing state-of-the-art deep learning architectures developed for pre-diagnostic screening of EEG are implemented and evaluated on the NMT, and referenced against baseline performance on the well-known Temple University Hospital EEG Abnormal Corpus. Generalization of deep learning based architectures across the NMT and the reference datasets is also investigated. The NMT dataset is being released to increase the diversity of EEG datasets and to overcome the scarcity of accurately annotated publicly available datasets for EEG research.

This paper presents a Convolutional Neural Networks (CNN) based Unsupervised Curriculum Learning approach for the recognition of water bodies to overcome the stated challenges for remote sensing based RGB imagery. The unsupervised nature of the presented algorithm eliminates the need for labelled training data. The problem is cast as a two class clustering problem (water and non-water), while clustering is done on deep features obtained by a pre-trained CNN. After initial clusters have been identified, representative samples from each cluster are chosen by the unsupervised curriculum learning algorithm for fine-tuning the feature extractor. The stated process is repeated iteratively until convergence. Three datasets have been used to evaluate the approach and show its effectiveness on varying scales: (i) SAT-6 dataset comprising high resolution aircraft images, (ii) Sentinel-2 of EuroSAT, comprising remote sensing images with low resolution, and (iii) PakSAT, a new dataset we created for this study. PakSAT is the first Pakistani Sentinel-2 dataset designed to classify water bodies of Pakistan. Extensive experiments on these datasets demonstrate the progressive learning behaviour of UCL and reported promising results of water classification on all three datasets. The obtained accuracies outperform the supervised methods in domain adaptation, demonstrating the effectiveness of the proposed algorithm.

Abstract Background With the advancement of powerful image processing and machine learning techniques, Computer Aided Diagnosis has become ever more prevalent in all fields of medicine including ophthalmology. These methods continue to provide reliable and standardized large scale screening of various image modalities to assist clinicians in identifying diseases. Since optic disc is the most important part of retinal fundus image for glaucoma detection, this paper proposes a two-stage framework that first detects and localizes optic disc and then classifies it into healthy or glaucomatous. Methods The first stage is based on Regions with Convolutional Neural Network (RCNN) and is responsible for localizing and extracting optic disc from a retinal fundus image while the second stage uses Deep Convolutional Neural Network to classify the extracted disc into healthy or glaucomatous. Unfortunately, none of the publicly available retinal fundus image datasets provides any bounding box ground truth required for disc localization. Therefore, in addition to the proposed solution, we also developed a rule-based semi-automatic ground truth generation method that provides necessary annotations for training RCNN based model for automated disc localization. Results The proposed method is evaluated on seven publicly available datasets for disc localization and on ORIGA dataset, which is the largest publicly available dataset with healthy and glaucoma labels, for glaucoma classification. The results of automatic localization mark new state-of-the-art on six datasets with accuracy reaching 100% on four of them. For glaucoma classification we achieved Area Under the Receiver Operating Characteristic Curve equal to 0.874 which is 2.7% relative improvement over the state-of-the-art results previously obtained for classification on ORIGA dataset. Conclusion Once trained on carefully annotated data, Deep Learning based methods for optic disc detection and localization are not only robust, accurate and fully automated but also eliminates the need for dataset-dependent heuristic algorithms. Our empirical evaluation of glaucoma classification on ORIGA reveals that reporting only Area Under the Curve, for datasets with class imbalance and without pre-defined train and test splits, does not portray true picture of the classifier’s performance and calls for additional performance metrics to substantiate the results.

We propose a novel hybrid approach that fuses traditional computer vision techniques with deep learning models to detect figures and formulas from document images. The proposed approach first fuses the different computer vision based image representations, i.e., color transform, connected component analysis, and distance transform, termed as Fi-Fo image representation. The Fi-Fo image representation is then fed to deep models for further refined representation-learning for detecting figures and formulas from document images. The proposed approach is evaluated on a publicly available ICDAR-2017 Page Object Detection (POD) dataset and its corrected version. It produces the state-of-the-art results for formula and figure detection in document images with an f1-score of 0.954 and 0.922, respectively. Ablation study results reveal that the Fi-Fo image representation helps in achieving superior performance in comparison to raw image representation. Results also establish that the hybrid approach helps deep models to learn more discriminating and refined features.

Following publication of the original article

Forged documents and counterfeit currency can be better detected with multispectral imaging in multiple color channels instead of the usual red, green and blue. However, multispectral cameras/scanners are expensive. We propose the construction of a low cost scanner designed to capture multispectral images of documents. A standard sheet-feed scanner was modified by disconnecting its internal light source and connecting an external multispectral light source comprising of narrow band light emitting diodes (LED). A document was scanned by illuminating the scanner light guide successively with different LEDs and capturing a scan of the document. The system costs less than a hundred dollars and is portable. It can potentially be used for applications in verification of questioned documents, checks, receipts and bank notes.

In-air signature is a new modality which is essential for user authentication and access control in noncontact mode and has been actively studied in recent years. However, it has been treated as a conventional online signature, which is essentially a 2D spatial representation. Notably, this modality bears a lot more potential due to an important hidden depth feature. Existing methods for in-air signature verification neither capture this unique depth feature explicitly nor fully explore its potential in verification. Moreover, these methods are based on heuristic approaches for fingertip or hand palm center detection, which are not feasible in practice. Inspired by the great progress in deep-learning-based hand pose estimation, we propose a real-time in-air signature acquisition method which estimates hand joint positions in 3D using a single depth image. The predicted 3D position of fingertip is recorded for each frame. We present four different implementations of a verification module, which are based on the extracted depth and spatial features. An ablation study was performed to explore the impact of the depth feature in particular. For matching, we employed the most commonly used multidimensional dynamic time warping (MD-DTW) algorithm. We created a new database which contains 600 signatures recorded from 15 different subjects. Extensive evaluations were performed on our database. Our method, called 3DAirSig, achieved an equal error rate (EER) of 0.46 %. Experiments showed that depth itself is an important feature, which is sufficient for in-air signature verification.

This paper evaluates the degree of saliency of texts in natural scenes using visual saliency models. A large scale scene image database with pixel level ground truth is created for this purpose. Using this scene image database and five state-of-the-art models, visual saliency maps that represent the degree of saliency of the objects are calculated. The receiver operating characteristic curve is employed in order to evaluate the saliency of scene texts, which is calculated by visual saliency models. A visualization of the distribution of scene texts and non-texts in the space constructed by three kinds of saliency maps, which are calculated using Itti's visual saliency model with intensity, color and orientation features, is given. This visualization of distribution indicates that text characters are more salient than their non-text neighbors, and can be captured from the background. Therefore, scene texts can be extracted from the scene images. With this in mind, a new visual saliency architecture, named hierarchical visual saliency model, is proposed. Hierarchical visual saliency model is based on Itti's model and consists of two stages. In the first stage, Itti's model is used to calculate the saliency map, and Otsu's global thresholding algorithm is applied to extract the salient region that we are interested in. In the second stage, Itti's model is applied to the salient region to calculate the final saliency map. An experimental evaluation demonstrates that the proposed model outperforms Itti's model in terms of captured scene texts.

Gender score is the cognitive judgement of the degree of masculinity or femininity of a face which is considered to be a continuum. Gender scores have long been used in psychological studies to understand the complex psychosocial relationships between people. Perceptual scores for gender and attractiveness have been employed for quality assessment and planning of cosmetic facial surgery. Various neurological disorders have been linked to the facial structure in general and the facial gender perception in particular. While, subjective gender scoring by human raters has been a tool of choice for psychological studies for many years, the process is both time and resource consuming. In this study, we investigate the geometric features used by the human cognitive system in perceiving the degree of masculinity/femininity of a 3D face. We then propose a mathematical model that can mimic the human gender perception. For our experiments, we obtained 3D face scans of 64 subjects using the 3dMDface scanner. The textureless 3D face scans of the subjects were then observed in different poses and assigned a gender score by 75 raters of a similar background. Our results suggest that the human cognitive system employs a combination of Euclidean and geodesic distances between biologically significant landmarks of the face for gender scoring. We propose a mathematical model that is able to automatically assign an objective gender score to a 3D face with a correlation of up to 0.895 with the human subjective scores.