The boehmite (AlOOH) sol derived from aluminium isopropoxide was synthesized by Yoldas method. The physicochemical process and phase transition of AlOOH were investigated by TG-DSC, N2 adsorption-desorption, XRD and TEM. The results showed that upon heating, AlOOH lost the physicsorbed water first, and then lost the chemisorbed water at 242 °C, and the dehydroxylation occurred at 361 °C. The BET surface area, total pore volume and pore diameter increased, since the waters were lost. The orthorhombic AlOOH transformed to the cubic γ-Al2O3 at 361 °C, the cubic η-Al2O3 appeared after 484 °C, the monoclinic θ-Al2O3 and rhombohedral α-Al2O3 appeared after 1026 °C, and it totally turned into α-Al2O3 at 1237 °C.

In this paper, a novel shot boundary detection algorithm is proposed that operates completely in the compressed domain using macroblock type information. While the proposed algorithm can save a lot of computation cost, extensive experiments support that the proposed algorithm also achieves superior performances. Such technique will provide a range of useful applications for on-line video content analysis, processing, and management. Experimental results show that the proposed shot boundary detection algorithm achieve a high recall, precision and F measure.