Recent advances in Artificial Intelligence (AI) have catalyzed the emergence of automated news generation systems, with growing research interest in semantic-driven, cross-modal approaches. This paper proposes INGF (Integrated News Generation Framework), an intelligent end-to-end system for generating image-text news that integrates eight journalistic writing styles and their corresponding prompts, curated through collaboration with journalism students. Our framework incorporates image filtering and quality optimization mechanisms, enabling semantic-aware cross-modal alignment to produce image-text news. We validate INGF on the dataset of Chinese image-text news from the Communication University of China (CUC) news platform. Comprehensive evaluations through quantitative metrics and a questionnaire survey (N = 101) demonstrate effectiveness of the eight designed news prompt types and readability of the generated image-text news.