Image Source Identification Using Convolutional Neural Networks in IoT Environment

Digital image forensics is a key branch of digital forensics that based on forensic analysis of image authenticity and image content. The advances in new techniques, such as smart devices, Internet of Things (IoT), artificial images, and social networks, make forensic image analysis play an increasing role in a wide range of criminal case investigation. This work focuses on image source identification by analysing both the fingerprints of digital devices and images in IoT environment. A new convolutional neural network (CNN) method is proposed to identify the source devices that token an image in social IoT environment. The experimental results show that the proposed method can effectively identify the source devices with high accuracy.

Coherent Image Source Modeling of Sound Fields in Long Spaces with a Sound-Absorbing Ceiling

Sound-absorbing boundaries can attenuate noise propagation in practical long spaces, but fast and accurate sound field modeling in this situation is still difficult. This paper presents a coherent image source model for simple yet accurate prediction of the sound field in long enclosures with a sound absorbing ceiling. In the proposed model, the reflections on the absorbent boundary are separated from those on reflective ones during evaluating reflection coefficients. The model is compared with the classic wave theory, an existing coherent image source model and a scale-model experiment. The results show that the proposed model provides remarkable accuracy advantage over the existing models yet is fast for sound prediction in long spaces.

Solving the boundary artifact for the enhanced deconvolution algorithm SUPPOSe applied to fluorescence microscopy

The SUPPOSe enhanced deconvolution algorithm relies in assuming that the image source can be described by an incoherent superposition of virtual point sources of equal intensity and finding the number and position of such virtual sources. In this work we describe the recent advances in the implementation of the method to gain resolution and remove artifacts due to the presence of fluorescent molecules close enough to the image frame boundary. The method was modified removing the invariant used before given by the product of the flux of the virtual sources times the number of virtual sources, and replacing it by a new invariant given by the total flux within the frame, thus allowing the location of virtual sources outside the frame but contributing to the signal inside the frame.