Efficient Encoding and Decoding Strategies for Information Visualization in Algorithm-Driven Visual Communication Design

Visual communication design often faces challenges such as information redundancy, high cognitive load, and instability in dynamic visualization, which hinder effective information transmission and user interaction. To address these issues, this study integrates multiple algorithm-driven optimization strategies, including contrast enhancement based on the Weber-Fechner law, visual saliency mapping via the Itti-Koch model, K-means clustering for color classification, color blindness-friendly palettes, force-directed and Voronoi-based layout algorithms, and smoothing techniques like exponential smoothing and Kalman filtering for dynamic data stabilization. An adaptive decoding framework based on Q-learning and cognitive load theory is also introduced to enhance interactive experience. Experimental results show that K-means optimized color schemes significantly improve recognition accuracy (up to 17.3%) and reduce response time, Voronoi layouts enhance node distribution and reduce task completion time by 27.2%, and Kalman filtering lowers data jitter to 5.8% while boosting trend recognition to 92.3%. The Q-learning-based interactive strategy reduces interaction delay by 35.7%, increases user satisfaction scores from 4.3 to 6.1, and shortens task completion time. These findings demonstrate that combining perceptual principles with machine learning techniques can effectively enhance the efficiency, accuracy, and user-friendliness of information visualization, though further exploration is needed to generalize the results across broader application scenarios and user profiles.