A Comprehensive Method for Flower Detection by DL: A Comparative Study between Multiple Statistical Models

Floweret recognition, a crucial aspect of plant identification and ecological research, plays a significant role in understanding biodiversity and ecosystem dynamics. This paper proposes a novel approach for floweret recognition leveraging advanced deep learning techniques. The proposed system utilizes convolutional neural networks (CNNs) to automatically extract hierarchical features from high-resolution images of flowerets, enabling accurate and efficient classification. The methodology involves a multi-stage process, starting with data preprocessing to enhance image quality and remove noise. Subsequently, a pre-trained CNN model, such as a modified ResNet50, VGG16, GoogleNet is fine-tuned using a diverse dataset of annotated floweret images. Transfer learning is employed to leverage the knowledge gained from a large dataset, allowing the model to generalize well to different floral species. In the end, our paper achieved 81.04% accuracy with the help of VGG-16, which is the highest compared to other algorithms.