Multisensor image fusion is a crucial research area aiming to enhance image clarity and comprehensibility by integrating information from multiple sensors. This paper presents a residual dense transformer (RDT) architecture for multisensor image fusion to address the challenges posed by the unique strengths and limitations of visual infrared (VIS), near-infrared (NIR), and long-wavelength infrared (LWIR) sensors. A comparative analysis is conducted with several state-of-the-art fusion methods using various objective evaluation indicators to asses the image fusion quality. We used a 313 triplet images collected from three datasets: TRICLOBS, MOFA, and MUDCAD, covering diverse environmental conditions such as foggy conditions and low illumination. Through the evaluation of the RDT and state-of-the-art fusion algorithms on this dataset, we observe that RDT achieve the best overall performance across multiple spectra image fusion. This work, thus, serves as a platform for developing and comparing new algorithms to deal with images from three sensors. which AIDS in the development of various applications such as object tracking, detection, and surveillance.