Wang Yun, Li Zhenbo, Li Fei, Mi Yang, Yue Jun
IEEE Trans Image Process. 2022;31:4994-5008. doi: 10.1109/TIP.2022.3191846. Epub 2022 Aug 1.
Representation learning is widely used to project high-dimensional data to low-dimensional subspace for feature extraction in image recognition tasks. However, many related methods barely explore the fuzziness and uncertainty between data classes. Besides, the classical unsupervised sparse constraint weakens the evaluation of feature importance and neglects the preservation of discriminant information during sparse representation. To solve these issues, a novel fuzzy discriminative block representation learning (FDBRL) algorithm is proposed for image feature extraction. FDBRL aims to enhance the discriminability of subspace by designing effective constraints for projection learning. Specifically, based on the label information and the fuzzy relation between data, we construct a fuzzy block weight matrix and embed it into the l norm regularization term to realize supervised sparse constraint for the representation learning. Next, the low-rank constraint is used to capture the inherent global structure information of data. Finally, we introduce a classification loss term with transformation matrix for joint optimization, such that the projection learning is not limited to number of classes, and the discriminative ability is further improved. Comprehensive experimental results on six benchmarks verify that our method achieves promising performance with other state-of-the-arts in both robustness and effectiveness.
表示学习在图像识别任务中被广泛用于将高维数据投影到低维子空间以进行特征提取。然而，许多相关方法几乎没有探索数据类之间的模糊性和不确定性。此外，经典的无监督稀疏约束削弱了对特征重要性的评估，并且在稀疏表示过程中忽略了判别信息的保留。为了解决这些问题，提出了一种新颖的模糊判别块表示学习（FDBRL）算法用于图像特征提取。FDBRL旨在通过为投影学习设计有效的约束来增强子空间的可判别性。具体而言，基于标签信息和数据之间的模糊关系，我们构造一个模糊块权重矩阵并将其嵌入到l范数正则化项中，以实现表示学习的监督稀疏约束。接下来，使用低秩约束来捕获数据的固有全局结构信息。最后，我们引入一个带有变换矩阵的分类损失项进行联合优化，使得投影学习不限于类的数量，并且判别能力进一步提高。在六个基准上的综合实验结果验证了我们的方法在鲁棒性和有效性方面与其他现有技术相比取得了有前景的性能。
