In this paper, we propose a novel method for producing image captions through the utilization of Generative Adversarial Networks (GANs) and Vision Transformers (ViTs) using our proposed Image Captioning Utilizing Transformer and GAN (ICTGAN) model. Here we use the efficient representation learning of the ViTs to improve the realistic image production of the GAN. Using textual features from the LSTM-based language model, our proposed model combines salient information extracted from images using ViTs. This merging of features is made possible using a self-attention mechanism, which enables the model to efficiently take in and process data from both textual and visual sources using the self-attention properties of the self-attention mechanism. We perform various tests on the MS COCO dataset as well as the Flickr30k dataset, which are popular benchmark datasets for image-captioning tasks, to verify the effectiveness of our proposed model. The outcomes represent that, on this dataset, our algorithm outperforms other approaches in terms of relevance, diversity, and caption quality. With this, our model is robust to changes in the content and style of the images, demonstrating its excellent generalization skills. We also explain the benefits of our method, which include better visual–textual alignment, better caption coherence, and better handling of complicated scenarios. All things considered, our work represents a significant step forward in the field of picture caption creation, offering a complete solution that leverages the complementary advantages of GANs and ViT-based self-attention models. This work pushes the limits of what is currently possible in image caption generation, creating a new standard in the industry.
In recent years, the agricultural sector has increasingly embraced advanced technologies to tackle food security and animal health issues. Poultry farming, a crucial part of agriculture, faces significant challenges from diseases affecting poultry health and economic sustainability. This project employs Convolutional Neural Networks (CNNs), a form of deep learning, to enhance poultry disease prediction accuracy, using chicken diseases as a case study. CNNs have revolutionised various fields, including disease prediction, by extracting meaningful patterns from data like images. This project leverages CNNs to analyse a diverse dataset of chicken disease images, creating a robust prediction model. The process involves compiling an extensive dataset of high-resolution chicken disease images, designing a CNN architecture with convolutional and pooling layers, and exploring transfer learning from pre-trained models. Rigorous training, validation, hyperparameter tuning, and data augmentation ensure model reliability. The project’s goals are twofold: demonstrating the feasibility of using CNNs for poultry disease prediction and offering a comprehensive poultry disease prediction framework. The latter could enable early disease detection and target interventions, reducing economic losses and enhancing food security. The proposed model achieved an overall accuracy of 96.5% and an F1 score of 96.8% respectively, on the tested dataset of poultry disease, indicating its high performance in poultry disease prediction.
1 May 2015 in Nigerian Journal of Mathematics and Applications
Nowadays, the use of information technology has taken over so many sectors. The survivals of many organizations are based on the information that is received or sent out. Due to the dependency on an unsecure communication network, some information that is confidential to the organization is revealed to an unauthorized user and this has necessitated the need for data security. This research work therefore made use of two algorithms for data cryptography to produce a more secure hybrid system. To achieve this aim, Advanced Encryption Scheme algorithm is used to encrypt a message and the secret key of Advanced Encryption is in turn encrypted using Ron Rivest Adi Shamir Leonard Adleman algorithm. This does not only produce a more secure method of data encryption, but also help to solve the problem of key sharing. The result shows that the hybrid Cryptographic sys tem is able to prevent data from unauthorised user.