This dataset was obtained in partnership with The Biological Dosimetry Laboratory(CRCN-NE) and The Computing Department (DC) of the Rural Federal University of Pernambuco (UFRPE), both located in the city of Recife - Brazil. Here are 74 images of colored metaphases acquired with a Leica microscope. The labeling of these metaphases will also be found, in this case, their expected results. In addition, we used the adaptive thresholding algorithm on this dataset, the result is also found in this file. So, in total are 74 images of colored metaphases and 74 images of segmented metaphases. Finally, cuts of regions of chromosomes and non-chromosomes will be found for both the color image and the segmented image. Being present in the same quantity for both distributions, since they were obtained by the same technique. Cuttings of chromosome regions occurred using the Python OpenCv library, which swept the labeled image and captured the regions concerning the chromosomes. With these coordinates it was possible to perform the mapping for the colored and segmented images, making the extraction of these areas of interest. The cuts of the regions of non-chromosomes occurred with the aid of the LabelMe tool, in it was carried out the marking of 46 random points in the image. With this tool, it is possible to obtain an xml with the coordinates of these markings, which were used to extract the areas of interest.