High resolution remote sensing data can provide worldwide images rapidly contrasted with conventional strategies for information accumulation. Therefore tiny objects like cars can be easily detected. Automatic vehicles enumeration research domain plays an important role in various applications including traffic monitoring and management. In this paper, we propose autonomous vehicle detection and classification approach in highway environment. Proposed approach consists mainly from three stages: (i) first, preprocessing operations are applied in order to eliminate noisy objects including soil, vegetation, water. (ii) Then, built-up area index is utilized to detect and delineate road networks. (iii) Finally, Multi-thresholding segmentation is implemented, resulting in vehicle detection and classification, where detected vehicles are classified into cars and trucks. Quality percentage assessment is carried over different study areas, illustrating the great efficiency of the proposed approach especially in highway environment.