一种基于DCT和SVM的人脸识别新方法

文中提出了一种新的人脸识别方法,该方法采用DCT提取人脸特征,并采用SVM对该特征进行分类识别。基于该方法,对ORL人脸库进行分类识别,仅用28个特征平均识别率达到97.4%。仿真结果表明,该方法显著地降低了特征维数和计算复杂度,明显提高了特征的可辨别能力,而且,SVM可以有效地提高分类器的分类和推广能力。     

Institutional parameters that condition farmer–herder conflicts in Tivland of Benue State,Nigeria

ABSTRACT

This conceptual review examines institutional parameters that underpin farmer–herder conflicts in Tivland of Benue State, Nigeria. Anchored on the theory of New Institutionalism, it argues that tensions and conflicts between Tiv-farmers and the Fulani-herdsmen is occasioned largely by the disarticulation of the traditional institutional norms/rules of interaction by formal political institutions expressed in land use laws, West African sub-regional protocol, human rights provision of the Nigerian constitution and the Benue State law on anti-grazing. In these contrasting institutional forms, the formal political institution is prioritised by the state while farmers and herders are divided about the institutional basis of their coexistence. Farmers who host the herdsmen demonstrate preference for traditional norms of land ownership and control while the herdsmen incline towards new political/legal instruments which they interpret, purports to give them access to grazing lands and hence an alibi for rejecting institutions that contradict their claim to access spaces for pastoralism. Thus, persisting claims to the exercise of rights to freedom of movement and access to grazing land as opposed to claims to traditional rights of ownership and control of ancestral lands explain the origin and continuity of the farmer–herder conflicts in Benue state, Nigeria.