The attribute reduction in the formal context is an important topic of formal concept analysis. Several kinds of attribute reductions in formal context have been put forward. This paper is devoted to the study of the reduction in formal context based on objects-induced three-way concept lattices. A new approach for granular reduction is proposed by using discernibility attributes of the objects. In this new approach of granular reduction, the objects-induced three-way concept lattices needs not to be constructed. Furthermore, it is proved that objects-induced three-way concept lattice based granular reduction and rough set based classification reduction are equivalent.