Climate change is a global problem which all countries are affected. The problem’s size may be depending on geography. A change of climate which is attributed directly or indirectly to human activity that alters the composition of global atmosphere which is in addition to natural climate variability observed over comparable time periods and it’s also directly related to global warming. Global warming could lead to higher rates of skin cancer by increasing the effects of the sun's rays. Skin cancer is the main cancer anxiety of the world when global warming’s impact. The skin cancer people get most is melanoma. Though melanoma is the most common, it is also the most curable and can even be prevented. Early discovery is very important. If the moles that are doing weird things such as changing size, shape, and color will become to the malignant melanoma. Malignant melanoma is a kind of cancer, which is one of the most deadly to skin cancer in the world. Dermoscopy, it is useful gadget used by dermatologists to assist with inspecting the skin. Thereby, the dermoscopy images were used by a dermatologist to analysis the skin lesions which were extensively used for non-invasive technique. Nevertheless, the diagnostic precision may gain actual incorrect results based on dermoscopy images due to inexperienced skill. Therefore, the advancement of computerized process would be beneficial tools for diagnosing skin cancers. The analysis of dermoscopy images based on computerized system is to monitor the border of the skin lesions. The ABCD criteria, the border detection of the skin lesions which is according to the Asymmetry, Border, Color, and Differential structure will be evaluated in the first step. Automatic interesting region of dermoscopy images is very appealing to investigate the border lesions. Thus, the border detection of dermoscopy image is an important part to help physicians for the purposes of diagnosing dermoscopy images as the skin lesions in malignant melanoma. In this paper, we propose a new technique to locate the skin lesion. The technique comprises of two parts; image pre-processing and image segmentation. Pre-processing method as the first part is used to remove some unwanted as a noise. In the second part, we proffer the approximate localization method as a new technique to detect the border of the skin lesions. The border detection results are collated with clinically ground truth, and assessed in terms of the percentage border error.