Abstract: To characterize the structures and properties of samples in the analysis of experimental data of Small-Angle Neutron Scattering (SANS), a physical model must be selected corresponding to each sample for iterative fitting. However, the conventional method of model selection is primarily based on manual experience, which has a high threshold for analysis and low accuracy. Furthermore, the automated selection of physical models based on standard neural networks face challenges such as the lack of local image features, large intra-class differences, and small inter-class differences. This paper proposes a Bimodal Feature Fusion Convolutional Neural Network (BFF-CNN) model to mitigate these issues. Initially, a physically informed Fourier-Bessel Transform (FBT) is deployed to extract global structural information from scattering images. Then, the original and FBT-transformed images are fed into two subnetworks for feature extraction and fusion, enhancing the overall feature representation capability of the neural network. A Restricted Softmax (R-Softmax) loss function is implemented, adding a penalty term to the original Softmax loss function for limiting the probability of input samples being assigned to incorrect classes. This alleviates the vanishing gradient problem when the Softmax loss approaches zero, thereby improving the convergence speed. Experimental results obtained using a self-built SANS image dataset show that the BFF-CNN significantly improves the prediction accuracy and average recall as compared to models such as the Residual Network (ResNet)-18 and PMG. Using the joint learning strategy of R-Softmax and center loss functions, the prediction accuracy and recall has improved by 5.4 and 10.5 percentage points, respectively, as compared to the case using only the Softmax loss function, demonstrating good classification performance for SANS data.

Key words: Small-Angle Neutron Scattering(SANS), physics model, neural network, Fourier-Bessel Transform(FBT), feature fusion

摘要： 小角中子散射(SANS)实验数据的分析过程需要科研人员选择样品对应的物理模型进行迭代拟合来表征样品的结构和属性。目前选择物理模型的方法大多是基于人工经验,分析门槛高、准确率较低。基于标准神经网络模型的小角中子散射实验样品物理模型自动化筛选方法面临着图像缺乏局部特征、类内差异大、类间差异小等问题。设计双模态特征融合卷积神经网络(BFF-CNN)模型,先引入物理感知的傅里叶-贝塞尔变换(FBT)来提取散射图像的全局结构信息,再将原始图像与FBT变换图像通过两个子网络分别进行特征提取与特征融合,以提升神经网络整体的特征表达能力。提出受限Softmax(R-Softmax)损失函数,通过在原生Softmax损失函数的基础上添加惩罚项来限制输入样本被分配到非本真类的概率,可在Softmax损失接近0时缓解梯度的消失问题,进而提高收敛速度。在自建的小角中子散射图像数据集上的实验结果表明,BFF-CNN的预测准确率和平均召回率相比于ResNet-18、PMG等模型提升幅度较大,采用R-Softmax与中心损失函数的联合学习策略后的预测准确率和召回率相比只采用Softmax损失函数提升了5.4和10.5个百分点,具有较好的小角中子散射图像分类效果。

关键词: 小角中子散射, 物理模型, 神经网络, 傅里叶-贝塞尔变换, 特征融合

CLC Number: 

• TP311.5