Graph Feature Extraction is the task to collect any important feature on the graph image so the extracted data can be further processed. In this paper, it is assumed that the graph consists only one single curve. The extracted features here are graph title, axis title, axis values, legend, scales, and curve values.To find the Region Of Interest (ROI) of the graph image, pixel projection is adopted. This is necessary to find the possible information location in the graph image. After the location has been located, then the alphanumeric data such as graph title, axis texts, and the legend of the graph are processed by the OCR (Optical Character Recognition).The values of the graph line is extracted by performing the scale calculation, and then, using the scale, the line pixel position is calculated to find the numeric data of the graph.The extracted data can be used for further processing such as graph interpretation system, pattern recognition, pattern abstraction, automatic graph reader system, etc. The data also can be used to analyze the graph when the graph came in the image form and is needed to be converted into numeric form.