http://localhost:8081/jspui/handle/123456789/15074 2026-10-21T22:23:31Z http://localhost:8081/jspui/handle/123456789/21597 Title: Enhancing Few-shot Semantic Segmentation Technique for Medical Image Segmentation Authors: Bagde, Abhishek Abstract: When confronted with new, unseen semantic classes, fully-supervised deep learning seg mentation models suffer, as their fine-tuning necessitates enormous quantities of data with annotation. Few-shot semantic segmentation (FSS) offers a solution by enabling the seg mentation of previously unseen semantic classes by using only a few labelled examples and without any reliance on fine-tuning. Existing state-of-the-art FSS approaches primar ily focus on segmenting natural imageset and rely on abundant annotated training data for effective generalization to testing classes that are unseen. Due to the above reasons, this training technique is not practical in clinical imaging situations with scarce annotations. To overcome the difficulties mentioned above, our work proposes enhancement in the exist ing self-supervised FSS framework for medical images called SSL-ALPNet; this framework eliminates the need for annotations which were required in the training phase and leverages superpixel-based pseudo-labels that provide supervision signals and also includes an adap tive local prototype pooling module integrated into the prototypical networks for improvis ing segmentation accuracy. The efficacy ofourmodelisdemonstratedinorgansegmentation using Abdominal-MRI images. Experimental results show that our model, which is an en hancement made to the SSL-ALPNet [1] approach, achieves better segmentation accuracy as compared to other standard Few-Shot-Segmentation methods whose training is based on a manually annotated dataset. 2023-06-01T00:00:00Z http://localhost:8081/jspui/handle/123456789/21596 Title: Hybrid Model for Impact Analysis of Climate Change on Droughts in Maharashtra State Authors: Gujar, Ameya Abstract: Droughts are prolonged periods of dry weather that have become more frequent and severe due to climate change and global warming. It can have devastating effects on agriculture, water resources, and ecosystems. Hence, a framework for the prediction of droughts is necessary for mitigating its impact, as it enables authorities to prepare and respond effec tively. This paper presents a hybrid model comprised of the Convolutional Neural Network and Gated Recurrent Units (called CNN-GRU) to predict the Standardized Precipitation Evapotranspiration Index (SPEI), which is used to measure drought intensity. We used India Meteorological Department (IMD) rainfall and temperature data of Maharashtra state of India during the years 1960-2021 as the historical dataset. Whereas, for the future pro jections, we used the Coupled Model Intercomparison Project Phase 6 (CMIP6) dataset of the same region during the years 2015-2100 for different Shared Socioeconomic Pathways (SSP) scenarios. Both these datasets include the daily precipitation, minimum temperature and maximum temperature values. The proposed model is trained and validated using IMD dataset and the final evaluation of its ability to predict the future droughts is conducted on the CMIP6 dataset. We confirmed that it outperforms in terms of mean squared error, mean absolute error, and root mean squared error over both the IMD and the CMIP6 datasets based on the comparative study with the existing deep learning models. 2023-06-01T00:00:00Z http://localhost:8081/jspui/handle/123456789/21595 Title: Impact Analysis of Data Augmentation on Neural Network Models Authors: Bhardwaj, Ankit Abstract: Data augmentation is a widely employed technique in deep learning to expand the training dataset artificially[1]. Its effectiveness in enhancing neural network models’ performance across computer vision and natural language processing tasks has been widely demonstrated[2]. This thesis presents a comprehensive study on the impact of various data augmentation techniques on neural network models. Experimentswereconductedonimagedataset, evaluatingdifferentneu ral network architectures with and without dataaugmentation. Theresultsreveal significant improvements in model accuracy and robustness due to dataaugmen tation. However, the choice of augmentation technique and parameters is crucial. experiments on YOLO[3] and Faster RCNN[4] object detection models demon strate notable performance enhancements with data augmentation. These find ings underscore the efficacy of data augmentation in improving deep learning model performance, particularly for object detection tasks. 2023-06-01T00:00:00Z http://localhost:8081/jspui/handle/123456789/21594 Title: Opcode Analysis for Android Malware Detection using Learning Techniques Authors: Anshul Abstract: The Android operating system’s open-source nature and growing popularity have attracted malware developers to target mobile devices. Android malware frequently obtains sensitive data and information, including banking passwords, social media profiles, contacts, SMS, etc. Security researchers developed Android malware de tection methods based on static and dynamic analysis of Android apps. The major ity of these methods, however, are susceptible to false positive and false negative detections. Obfuscated malware can easily circumvent machine learning-based de tection techniques using code encryption and reflection. Therefore, I intend to ad dress the issue of code obscurity in Android malware detection. I utilized Dalvik machine-level opcodes to detect Android malware. Android applications are pack aged in an APK file that contains an AndroidManifest.xml file, dalvik executable code (classes.dex), and other resource files such as images, videos, icons, audio, etc. Malware detection models based on AndroidManifest features can be bypassed by introducing extra permissions and intents. Obfuscated malware has the potential of circumventing API call-based malware detection models. I designed and imple mented an Android malware detection model using instruction-level opcodes ex tracted from dex files. First, we use APKTool to decompress the APK files to obtain classes.dex files. Next, the opcodes are extracted from the dex files. The Android op erating system supports 218 opcodes in total. Furthermore, I generated the sequence of opcodes and applied one-hot encoding to transform the sequences into feature vectors. Deep learning models such as CNN, RNN, and LSTM have proven to be accurate in classifying text and images into various classes. I argued that convert ing the malware classification problem into a text classification problem can detect malware more precisely. Therefore, I constructed the sequence of opcodes and em ploy deep learning to classify any unknown Android app as malicious or benign. I used a dataset of 4,000 malware and benign applications to train and evaluate our deep learning-based detection model. The experimental results indicate that our proposed detection model is immune to code obfuscation and encryption and can detect malware with a 94.82% accuracy. 2023-05-01T00:00:00Z