Post-mortem livor mortis represents the blood accumulation in body areas underneath gravity which aids forensic examiners to determine both the period and conditions surrounding death. The latest machine learning developments created alternative methods for automated forensic recognition and classification of livor mortis patterns. A survey employs machine learning algorithms to recognize patterns in livor mortis data along with classification techniques along with performance data collection through datasets. This study examines three types of feature extraction methods which include texture analysis, color histograms and shape descriptors. Moreover, it evaluates three machine learning algorithms namely Support Vector Machines (SVM), Convolutional Neural Networks (CNNs) and Random Forests (RF). The study investigates the significance of deep learning models should be utilized for better livor mortis pattern recognition accuracy. The survey identifies dataset limits as well as algorithm interpretability and feature variation as future research areas along with their proposed directions. The review delivers complete foundations for using machine learning methods in forensic tasks including support for both secure and rapid death investigation procedures.