基于字典学习的煤与矸石图像特征识别方法

为了提升采煤过程中煤与矸石图像的自动识别效率,解决现有图像特征提取方式的瓶颈,根据相应的机器视觉原理,提出了利用字典学习算法对煤与矸石的图像进行特征提取的方法。字典学习算法经过煤与矸石图像的预处理之后,在字典初始化与更新时采用随机选择的方式,调整到最优的字典列数、稀疏误差和稀疏度等参数,最后利用支持向量机进行识别。实验表明:利用字典学习算法可以有效地表达煤与矸石的图像特征,自动识别效果较好,为采矿过程中煤与矸石的自动分选提供了一种新的思路和方法。     

Role of Trust Transfer in E-Commerce Acceptance

Lack of trust is one of the most important obstructions to consumers’ acceptance of e-commerce. With the increase of companies going online from offline, trust transfer has been found between offline presence and online presence. By applying the entitativity theories, this study empirically tested the trust transference from offline presence to online presence and addressed the underlying reasons through a questionnaire investigation on the customers of Lianhua supermarkets and Lianhua OK online supermarkets. Process integration was used as a useful means to promote this transfer process. For a click-and-mortar company such as Lianhua Group, its established offline trust is one of the major sources contributing to its initial online trust building. The results from this study suggest a new way to enhance e-commerce acceptance.