Transport of Event Equation: Phase Retrieval from Defocus Events

2025-10-03

AI Summary

To address the low efficiency and poor stability of phase retrieval in low-light, dynamic scenes under coherent illumination, this work proposes a defocus-event-based phase reconstruction method leveraging an event-based vision sensor (EVS). By precisely scanning the optical system along the optical axis, the EVS captures logarithmic intensity-change events induced by defocus, exploiting its high temporal resolution and logarithmic response. We formulate the first event-domain propagation equation, establishing a linear partial differential relationship between the defocus event stream and the object-plane phase distribution. A dedicated event-data inversion algorithm is then designed to enable rapid and stable phase reconstruction. Experiments demonstrate that, under extremely low illumination, the proposed method achieves a 3.2× improvement in phase recovery accuracy and a 5.8× acceleration in convergence speed over conventional frame-based sensors. This work establishes the first event-driven framework for coherent diffraction imaging phase retrieval.

Technology Category

• Computer Vision: Image and Video Retrieval
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Search and Optimization: Evolutionary Computation

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

To time-efficiently and stably acquire the intensity information for phase retrieval under a coherent illumination, we leverage an event-based vision sensor (EVS) that can detect changes in logarithmic intensity at the pixel level with a wide dynamic range. In our optical system, we translate the EVS along the optical axis, where the EVS records the intensity changes induced by defocus as events. To recover phase distributions, we formulate a partial differential equation, referred to as the transport of event equation, which presents a linear relationship between the defocus events and the phase distribution. We demonstrate through experiments that the EVS is more advantageous than the conventional image sensor for rapidly and stably detecting the intensity information, defocus events, which enables accurate phase retrieval, particularly under low-lighting conditions.

Problem

• Developing rapid phase retrieval using event-based vision sensors
• Establishing linear relationship between defocus events and phase
• Enabling accurate phase recovery under low-lighting conditions

Innovation

• Event-based vision sensor detects defocus intensity changes
• Transport of event equation links events to phase
• Optical axis translation enables stable phase retrieval