Roads have become an important element in city planning. By having a good quality road, a city will be able to provide better access and accommodation which will support the economy and eventually the country development. However, problems arise when the roads are easily cracked and holed. Maintenance and construction must become a priority. Periodical checking is one of the important strategies to supervise, maintain, and monitor the roads ‘conditions. Nevertheless, in checking the roads' condition, the employees are still using the manual method, in which they directly come to the roads and check them manually. This manual method is considered inefficient in the terms of duration, the numbers of workers, and accuracy because it is still using merely human power. Detection on roads cracks can be detected by using digital image processing. First of all, the figure of the roads can be taken by using a drone (unmanned aerial vehicle) and analyzed in a laboratory to determine the cracked roads which should be fixed immediately. This study is aimed to develop segmenting images on the road in order to facilitate the identification of cracks on the road. This research described the image segmentation of lacunarity method. The new method proposed in this research utilized the concept of mapping lacunarity values to separate cracked area on the road image. This present study aimed at conducting cracked roads image segmentation and detection by using a drone. The new method which was coined out in this study implemented the concept of Lacuranity mapping values to separate the cracked areas on roads images. Extracted texture feature through Lacuranity values on the images was really helpful in detecting and segmenting the cracks, primarily on the images with the same gray scale between the cracked areas and the road.