Samiul Islam

Face recognition technology has advanced significantly, yet many existing systems struggle to recognize new faces without large labeled datasets. This study introduces a custom CLIP-based framework that integrates Zero-Shot Learning (ZSL) for face recognition without extensive retraining. By leveraging visual-text alignment, the model maps facial features to descriptive text representations in a shared embedding space, ensuring adaptability to unseen identities. A projection layer is incorporated post-feature extraction to enhance alignment, improving recognition accuracy. Additionally, contrastive learning optimizes image-text relationships, allowing effective generalization to unseen classes while preserving zero-shot capability. Extensive experiments validated the framework’s effectiveness, achieving a validation accuracy of 0.9621, outperforming state-of-the-art models. In generalization tests on unseen classes, it attained 0.78 accuracy with high ROC-AUC scores, surpassing traditional face recognition models. The framework was also tested under varying image conditions, such as different resolutions and viewing angles, and maintained stable accuracy, proving its robustness in real-world scenarios. These results highlight its scalability and efficiency, making it well-suited for applications requiring adaptability to unseen identities. The proposed framework provides a practical, high-performance solution for modern face recognition, addressing key limitations in existing systems while ensuring adaptability to diverse and dynamic environments.

The Software-Defined Network (SDN) is a next-generation network that uses OpenFlow to decouple the control plane from the data plane of forwarding devices. Other protocols for southbound interfaces include ForCES and POF. However, some security issues might be in action on the SDN, so that attackers can take control of the SDN control plane. Since live video calling, QoS control, high bandwidth needs, and resource management are inevitable in any SDN/Software-Defined Cellular Network (SDCN), traffic monitoring is an integral approach for safeguarding against DDoS, heavy hitters, and superspreaders. In such a scenario, SDN traffic measurement comes into action. Thus, we survey SDN traffic measurement solutions to assess how these solutions can make a secure, efficient, and robust SDN/SDCN architecture. This research classifies SDN traffic measurement solutions according to network application behavior and compares several ML approaches. Furthermore, we find out the challenges related to SDN/SDCN traffic measurement and the future scope of research, which will guide the design and development of more advanced traffic measurement solutions for a scalable, heterogeneous, hierarchical, and widely deployed SDN/SDCN architecture. In more detail, we list different kinds of practical machine learning (ML) approaches to analyze how we can improve traffic measurement performances. We conclude that using ML in SDN traffic measurement solutions will help secure SDNs/SDCNs in complementary ways.

A common task in land-cover classification is water body extraction, wherein each pixel in an image is labelled as either water or background. Water body detection is integral to the field of urban hydrology, with applications ranging from early flood warning to water resource management. Although traditional index-based methods such as the Normalized Difference Water Index (NDWI) and the Modified Normalized Difference Water Index (MNDWI) have been used to detect water bodies for decades, deep convolutional neural networks (DCNNs) have recently demonstrated promising results. However, training these networks requires access to large quantities of high-quality and accurately labelled data, which is often lacking in the field of remotely sensed imagery. Another challenge stems from the fact that the category of interest typically occupies only a small portion of an image and is thus grossly underrepresented in the data. We propose a novel approach to data augmentation—pixelwise category transplantation (PCT)—as a potential solution to both of these problems. Experimental results demonstrate PCT’s ability to improve performance on a variety of models and datasets, achieving an average improvement of 0.749 mean intersection over union (mIoU). Moreover, PCT enables us to outperform the previous high score achieved on the same dataset without introducing a new model architecture. We also explore the suitability of several state-of-the-art segmentation models and loss functions on the task of water body extraction. Finally, we address the shortcomings of previous works by assessing each model on RGB, NIR, and multispectral features to ascertain the relative advantages of each approach. In particular, we find a significant benefit to the inclusion of multispectral bands, with such methods outperforming visible-spectrum models by an average of 4.193 mIoU.

In satellite remote sensing applications, waterbody segmentation plays an essential role in mapping and monitoring the dynamics of surface water. Satellite image segmentation—examining a relevant sensor data spectrum and identifying the regions of interests to obtain improved performance—is a fundamental step in satellite data analytics. Satellite image segmentation is challenging for a number of reasons, which include cloud interference, inadequate label data, low lighting and the presence of terrain. In recent years, Convolutional Neural Networks (CNNs), combined with (satellite captured) multispectral image segmentation techniques, have led to promising advances in related research. However, ensuring sufficient image resolution, maintaining class balance to achieve prediction quality and reducing the computational overhead of the deep neural architecture are still open to research due to the sophisticated CNN hierarchical architectures. To address these issues, we propose a number of methods: a multi-channel Data-Fusion Module (DFM), Neural Adaptive Patch (NAP) augmentation algorithm and re-weight class balancing (implemented in our PHR-CB experimental setup). We integrated these techniques into our novel Fusion Adaptive Patch Network (FAPNET). Our dataset is the Sentinel-1 SAR microwave signal, used in the Microsoft Artificial Intelligence for Earth competition, so that we can compare our results with the top scores in the competition. In order to validate our approach, we designed four experimental setups and in each setup, we compared our results with the popular image segmentation models UNET, VNET, DNCNN, UNET++, U2NET, ATTUNET, FPN and LINKNET. The comparisons demonstrate that our PHR-CB setup, with class balance, generates the best performance for all models in general and our FAPNET approach outperforms relative works. FAPNET successfully detected the salient features from the satellite images. FAPNET with a MeanIoU score of 87.06% outperforms the state-of-the-art UNET, which has a score of 79.54%. In addition, FAPNET has a shorter training time than other models, comparable to that of UNET (6.77 min for 5 epochs). Qualitative analysis also reveals that our FAPNET model successfully distinguishes micro waterbodies better than existing models. FAPNET is more robust to low lighting, cloud and weather fluctuations and can also be used in RGB images. Our proposed method is lightweight, computationally inexpensive, robust and simple to deploy in industrial applications. Our research findings show that flood-water mapping is more accurate when using SAR signals than RGB images. Our FAPNET architecture, having less parameters than UNET, can distinguish micro waterbodies accurately with shorter training time.

With the advancements in remote sensing, satellite images have become a popular data modality and are widely used in many applications. Among others, satellite images are used by surveyors and engineers for urban planning and asset management, e.g., for rapidly growing cities, urban sprawl, and informal settlement. A preliminary step in satellite image analysis is identifying the regions of interest or target objects in the scene and extracting them from irrelevant objects for further processing. This image segmentation task is challenging due to various factors, e.g., occlusions, lighting conditions, and noises. To address these issues, many researchers have proposed different approaches, e.g., UNET, which is regarded as a state-of-the-art method. However, it has high memory consumption, low accuracy, and poor quality prediction when applying satellite images. Thus, we propose a multi-channel decomposition semantic segmentation method, MNET, for object classification. We used the dataset obtained by MBRSC satellites and divided the scene objects into six classes (vegetation, water, road, building, land, unused). The study area is in Dubai. We compared our approach with UNET++, U2NET, and 2D-VNET. Experimental findings show that: (1) MNET outperforms others with a mIoU score of 79.4%, whereas UNET gives 75.1%, and (2) accurate patch generation is a crucial part of the model performance, as ignoring less informative patches during training increases the accuracy by 5%. In conclusion, our proposed method offers a lightweight structure, is a computationally less expensive model, and is simple to deploy in industry applications. In this work, we conducted an extensive analysis of domain transferability, class bias, and class co-occurrence. We also compared the best and worst predictions with a number of current methods. We believe that our MNET can also benefit other domains, such as in floodwater detection using Interferometric Synthetic Aperture Radar (InSAR) data, building extraction, and crop segmentation, due to the prediction accuracy and memory consumption efficiency of MNET.

The intent of the research work was to develop and validate a simple, selective and precise reversed phase ultra-high performance liquid chromatography (RP-UHPLC) method for the determination of ledipasvir. A forced degradation study was performed as per International Conference on Harmonisation (ICH) guidelines Q1A (R2) and Q1B. Ledipasvir was found to be well separated from degradation products using an analytical C18 column (150 mm × 4.6 mm i.d., 5 μm particle size) with a ratio of mobile phase (75:25 v/v) consisting of methanol and 0.1% trifluoroacetic acid (TFA) kept at ambient temperature. The average retention time of ledipasvir was found 4.45 min at 254 nm wavelength with 1.6 mL/min isocratic flow rate and 10 µL injection volume. Linearity, accuracy, precision, sensitivity, robustness, and ruggedness were studied according to ICH guideline Q2 (R1) to validate the method. Then, this validated method was applied for forced degradation studies of ledipasvir. In conclusion, the developed method has been successfully used to study degradation behaviour of ledipasvir and may be useful to quantify the drug in different pharmaceutical dosage forms.

Vision loss happens due to diabetic retinopathy (DR) in severe stages. Thus, an automatic detection method applied to diagnose DR in an earlier phase may help medical doctors to make better decisions. DR is considered one of the main risks, leading to blindness. Computer-Aided Diagnosis systems play an essential role in detecting features in fundus images. Fundus images may include blood vessels, exudates, micro-aneurysm, hemorrhages, and neovascularization. In this paper, our model combines automatic detection for the diabetic retinopathy classification with localization methods depending on weakly-supervised learning. The model has four stages; in stage one, various preprocessing techniques are applied to smooth the data set. In stage two, the network had gotten deeply to the optic disk segment for eliminating any exudate's false prediction because the exudates had the same color pixel as the optic disk. In stage three, the network is fed through training data to classify each label. Finally, the layers of the convolution neural network are re-edited, and used to localize the impact of DR on the patient's eye. The framework tackles the matching technique between two essential concepts where the classification problem depends on the supervised learning method. While the localization problem was obtained by the weakly supervised method. An additional layer known as weakly supervised sensitive heat map (WSSH) was added to detect the ROI of the lesion at a test accuracy of 98.65%, while comparing with Class Activation Map that involved weakly supervised technology achieved 0.954. The main purpose is to learn a representation that collect the central localization of discriminative features in a retina image. CNN-WSSH model is able to highlight decisive features in a single forward pass for getting the best detection of lesions.