It is well known that a machine vision-based analysis of a dynamic scene, for example in the context of advanced driver assistance systems (ADAS), does require real-time processing capabilities. Therefore, the system used must be capable of performing both robust and ultrafast analyses. Machine vision in ADAS must fulfil the above requirements when dealing with a dynamically changing visual context (i.e. driving in darkness or in a foggy environment, etc). Among the various challenges related to the analysis of a dynamic scene, this paper focuses on contrast enhancement, which is a well-known basic operation to improve the visual quality of an image (dynamic or static) suffering from poor illumination. The key objective is to develop a systematic and fundamental concept for image contrast enhancement that should be robust despite a dynamic environment and that should fulfil the real-time constraints by ensuring an ultrafast analysis. It is demonstrated that the new approach developed in this paper is capable of fulfilling the expected requirements. The proposed approach combines the good features of the ‘coupled oscillators’-based signal processing paradigm with the good features of the ‘cellular neural network (CNN)’-based one. The first paradigm in this combination is the ‘master system’ and consists of a set of coupled nonlinear ordinary differential equations (ODEs) that are (a) the so-called ‘van der Pol oscillator’ and (b) the so-called ‘Duffing oscillator’. It is then implemented or realized on top of a ‘slave system’ platform consisting of a CNN-processors platform. An offline bifurcation analysis is used to find out, a priori, the windows of parameter settings in which the coupled oscillator system exhibits the best and most appropriate behaviours of interest for an optimal resulting image processing quality. In the frame of the extensive bifurcation analysis carried out, analytical formulae have been derived, which are capable of determining the various states of the system (e.g. quenching state, non-zero equilibrium state, oscillatory states, etc). Both equilibrium and oscillatory states of the system have been depicted. It could be shown that each of these states has a significant impact on the quality of the resulting image contrast enhancement. A benchmark has been considered, whereby a comparison is performed between the results’ quality obtained by traditional CNN-based processing, on the one hand, with those obtained by ‘coupled nonlinear oscillators’-based processing, on the other hand. Thus, the superiority of the latter approach in terms of ensuring a constantly high enhancement quality despite luminosity-related spatio-temporal scene dynamics has been demonstrated both analytically/conceptually and through various experiments. A key drawback of the latter approach is, however, the potentially huge and challenging computing effort necessary for solving the coupled nonlinear and highly stiff ODEs when one attempts to solve them numerically on ‘von Neumann’-type computing platforms; this is evidently difficult to realize in real time for frame rates necessary in ADAS. On the other side, a major drawback of the traditional CNN-based image processing is the practical inability to adjust/recalculate templates in real time in the face of a dynamic scene, that is, in the presence of input images experiencing visibility- and/or lighting-related spatio-temporal dynamics. Nevertheless, CNN has the great advantage of ensuring ultrafast processing due to its inherent parallel processing and ‘analogue computing’ nature. Thus, the two approaches appear to complement each other very well. Therefore, the novel hybrid approach does integrate both schemes in an efficient way, that is ‘coupled nonlinear oscillators’-based image processing is the main processing scheme that is realized on top of a CNN-processors framework. The hybrid approach does evidently overcome the mentioned key practical problems faced by both the original approaches.
Read full abstract