Abstract:Image captioning is a challenging task involving generating descriptive sentences to describe images. The application of semantic concepts to automatically annotate images has made significant progress. However, the now available frameworks have apparent limitations, particularly in concept detection. Incomplete labelling due to biased annotations, using synonyms in training captions, and the enormous gap between positive and negative thought samples contribute to the problem. Incomplete labelling is a result of biased annotations. The captioning frameworks that are now in use are inadequate and create a barrier to accurate image captioning. Unequal sample occurrences and missing training captions negatively affect the model's potential to develop rich and varied descriptions of images. Inadequate sample occurrences and missing training captions also contribute to insufficient idea generation. To circumvent these limitations, a novel approach has been designed to automatically generate images using Weighted Stacked Generative Adversarial Network (WSGAN). With the help of this boost, the uneven distribution of concepts is intended to be rectified, thereby expanding the breadth of the horizons covered by the training set. The proposed approach utilizes a WSGAN in conjunction with a Gated Recurrent Units (GRU)–based Deep Learning (DL) model and a Visual Attention Mechanism (VAM)–based DL model. The purpose of the GRU-VAM model is to enable the generation of text captions for images. To train the model, combining the MS COCO dataset with a wide variety of original and machine-generated image datasets in numerous permutations is necessary. The WSGAN-generated images correct the imbalance and incompleteness in the training dataset, which boosts the model's ability to capture a wider variety of thoughts. During testing and evaluation, the proposed WSGAN- GRU-VAM demonstrates significant enhancements in image captioning metrics compared to existing models. WSGAN-GRU-VAM is superior to other well-known image captioning algorithms such as EnsCaption, Fast RF-UIC, RAGAN, and SAT-GPT-3 in terms of its performance across various essential parameters. Increase in BLEU (8%), METEOR (7%), CIDEr (9%), and ROUGE-L (6%), on average, reflect the model's capacity to provide image captions with enhanced linguistic accuracy, relevance, and coherence.

Taylor African vulture optimization algorithm with hybrid deep convolution neural network for image captioning system

Adaptive Coati Optimization Enabled Deep CNN-based Image Captioning

Advanced Generative Deep Learning Techniques for Accurate Captioning of Images

An efficient automated image caption generation by the encoder decoder model

An image caption model based on attention mechanism and deep reinforcement learning

An Intelligent Image Caption Generator Model using Deep Learning

Advancing image captioning with V16HP1365 encoder and dual self-attention network

Avtmnet: Adaptive Visual-Text Merging Network for Image Captioning

Image Caption Generator Using Deep Learning

Recurrent Image Captioner: Describing Images with Spatial-Invariant Transformation and Attention Filtering

Enhancing Image Captioning Using Deep Convolutional Generative Adversarial Networks

GVA: guided visual attention approach for automatic image caption generation

Image Captioning using Deep Stacked LSTMs, Contextual Word Embeddings and Data Augmentation

Neuraltalk+: neural image captioning with visual assistance capabilities

D-CNN: A New model for Generating Image Captions with Text Extraction Using Deep Learning for Visually Challenged Individuals

Insights into Object Semantics: Leveraging Transformer Networks for Advanced Image Captioning

Automatic image captioning system based on augmentation and ranking mechanism

Synthesis of Vision and Language: Multifaceted Image Captioning Application

Dual visual align-cross attention-based image captioning transformer

An accurate generation of image captions for blind people using extended convolutional atom neural network

An improved African vultures optimization algorithm using different fitness functions for multi-level thresholding image segmentation