Shen Gang, Ma Wenjun, Zhai Wen, Lv Xuefei, Chen Guangyao, Tian Yonghong
Smart Tower Co., Ltd., Beijing 100089, China.
State Unclear Electric Power Planning Design & Research Institute Co., Ltd., Beijing 100095, China.
Entropy (Basel). 2025 Apr 18;27(4):436. doi: 10.3390/e27040436.
Biologically inspired retinal preprocessing improves visual perception by efficiently encoding and reducing entropy in images. In this study, we introduce a new saliency prediction framework that combines a retinal model with deep neural networks (DNNs) using information theory ideas. By mimicking the human retina, our method creates clearer saliency maps with lower entropy and supports efficient computation with DNNs by optimizing information flow and reducing redundancy. We treat saliency prediction as an information maximization problem, where important regions have high information and low local entropy. Tests on several benchmark datasets show that adding the retinal model boosts the performance of various bottom-up saliency prediction methods by better managing information and reducing uncertainty. We use metrics like mutual information and entropy to measure improvements in accuracy and efficiency. Our framework outperforms state-of-the-art models, producing saliency maps that closely match where people actually look. By combining neurobiological insights with information theory-using measures like Kullback-Leibler divergence and information gain-our method not only improves prediction accuracy but also offers a clear, quantitative understanding of saliency. This approach shows promise for future research that brings together neuroscience, entropy, and deep learning to enhance visual saliency prediction.
受生物启发的视网膜预处理通过有效编码和降低图像熵来改善视觉感知。在本研究中，我们引入了一种新的显著性预测框架，该框架利用信息论思想将视网膜模型与深度神经网络（DNN）相结合。通过模仿人类视网膜，我们的方法创建了具有更低熵的更清晰的显著性图，并通过优化信息流和减少冗余来支持DNN的高效计算。我们将显著性预测视为一个信息最大化问题，其中重要区域具有高信息和低局部熵。在几个基准数据集上的测试表明，添加视网膜模型通过更好地管理信息和减少不确定性，提高了各种自底向上显著性预测方法的性能。我们使用互信息和熵等指标来衡量准确性和效率的提高。我们的框架优于现有模型，生成的显著性图与人们实际注视的位置非常匹配。通过将神经生物学见解与信息论相结合——使用诸如库尔贝克-莱布勒散度和信息增益等度量——我们的方法不仅提高了预测准确性，还提供了对显著性的清晰、定量理解。这种方法为未来将神经科学、熵和深度学习结合起来以增强视觉显著性预测的研究展现出了前景。
