Multisensor image fusion using fuzzy logic for surveillance systems

Abstract

Multisensor image fusion has its effective utilization for surveillance. In this paper, we utilize a fuzzy logic approach to fuse images from different sensors, in order to enhance visualization for surveillance. With the help of fuzzy if-then rules and membership functions designed for the image data set, the fuzzy logic approach can model and combine the images to enhance the contrast of the fused image. Mamdani-type fuzzy inference system is used to combine the images. Subjective and objective image fusion performance measures are introduced to assess the performance of image fusion schemes. Experimental results show that the fuzzy logic approach is effective to fuse images from different sensors.

Улучшение Визуального Качества Изображений На Основе Инфракрасных Данных В Системах Видеонаблюдения

“…Если объекты корректно распознаны со всех источников изображений, то происходит их слияние для получения действительного представления сцены. Наиболее распространенными подходами слияния изображений на основе выделения объектов являются алгоритмы нечеткой логики [19], байесовская сеть [20] и алгоритмы машинного обучения [21].…”

Multi‐level fuzzy contourlet‐based image fusion for medical applications

“…Region (feature) fusion rules group image pixels to form contiguous regions, for example, objects and impose different fusion rules to each image region. This type of fusion rule requires the extraction of different features from the source data before features are merged together, which can be attributed to the inefficiencies faced by pixel‐based algorithms in cases where the salient features in images are larger than one pixel [8, 9]. In general, pixel‐based fusion algorithms concentrate on increasing image contrast whereas region‐based algorithms provide edge enhancement.…”

Performance assessment of neuro fuzzy based image fusion of satellite images

“…Image fusion method has been utilized in pronounced domains: medical image processing, satellite image processing, computer vision, involuntary change recognition, biometrics and armed solicitations. Multi-device image combination for investigation schemes deliberated where fuzzy method exploited for fusing images taken from various sensors, in order to improve conception for observation [2]. The source images decomposition by wavelet transform three consistency structures are mined and then a fuzzy instruction is utilized to combine wavelet factors from the two images conferring to the mined structures.…”