Cognitive Radio (CR) networks have received much attention as a solution to the spectrum inefficiency problem of current license-based regulatory management. In CR networks, Secondary User (SU) can use a spectrum vacancy that can be detected by either sensing-before-transmission or database access. However, sensing inaccuracy or long access time to database often becomes a major obstacle to timely detect the spectrum vacancy. In this paper, we develop a hybrid detection framework in multi-channel CR networks, where an SU can selectively sense a channel for spectrum vacancy by accessing the spectrum history of Markovian channels. We focus on the value of the channel history information offered by the Primary Provider (PP) of each channel, and consider a market for the information between multiple PPs and SU. We investigate the interplay between of the PPs and the SU through their pricing and buying decisions for the information, in the presence of sensing inaccuracy, i.e., false alarm and miss detection.