This book chapter presents an image processing tool developed to in-vitro human oocytes analysis required in fertilization processes to biological material classification. The application of the proposed tool allows a non-invasive method to define the quality of human oocytes before to be inseminated, enabling the possibility of selection the optimum one. Specific segmentation algorithms have been developed to specific biomedical images for identifying the cytoplasm area. The oocyte status (normal, medium and highly stressed state) has been defined in terms of a set of parameters extracted from the cytoplasm image. This status quantification allows advising the biomedical staff during oocyte selection, involving by the first time a useful metrics. Experimental results proof that 83% of cases analyzed match with the expert evaluation.