Oil palm is a highly efficient crop that can produce more oil per unit of land than any other type of oil seed. Palm oil it is in high demand, and its production can significantly contribute to a country’s economic growth. However, the traditional method of grading palm fruit is still prevalent in Malaysia, which requires skilled workers to classify the harvested fruit according to its ripeness. This approach can be costly and labour-intensive. Therefore, several studies have investigated automated palm fruit classification techniques that could reduce costs and labour in the industry. This paper provides a review of these studies, with a specific focus on vision-based classification techniques. The article discusses approaches based on image processing encompassing pre-processing, feature extraction and classification steps. The survey’s results indicate that there is a lack of technique to effectively address outdoor images, such as colour correction methods. Therefore, further research is necessary to develop a better segmentation and colour correction procedures. Overall, the findings of this study could help improve the efficiency and sustainability of palm oil production, thereby contributing to economic growth and environmental conservation.