Weak light image enhancement method based on multi-branch progressive deep network

• Main Authors: YU YOUJIANG, YUAN CHENG, CHEN XIAOGAI, LI MINQI, ZHANG KAIBING, LU JIAN
• Format: Patent
• Language: Chinese; English
• Published: 18.07.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention discloses a multi-branch progressive weak light image enhancement method. The method comprises the following steps: step 1, preparing data; step 2, building a multi-branch progressive enhancement model; 3, training the data prepared in the step 1 according to the structure built in the step 2 to obtain a trained multi-branch progressive enhancement model; and 4, performing data testing based on the trained multi-branch progressive enhancement model obtained in the step 3. According to the method, the problem of low-light image degradation is solved through the multi-branch enhancement networks with different scales. 本发明公开了一种基于多分支渐进式的弱光图像增强方法，包括如下步骤：步骤1，准备数据；步骤2，搭建多分支渐进式增强模型；步骤3，根据步骤2搭建的结构对步骤1准备的数据进行训练，得到训练好的多分支渐进式增强模型；步骤4，基于步骤3得到的训练好的多分支渐进式增强模型进行数据测试。本发明通过不同尺度的多分支增强网络克服了弱光图像退化问题。