An underwater disaster can be declared when a maritime accident occurred or when an aircraft is plunged into water area, be it ocean, sea or river. Nevertheless, handling of human remains in an underwater recovery operation is often a difficult and demanding task as working conditions may be challenging with poor to no visibility, location of remains at considerable depths and associated hazards from surrounding water. A case of the recent helicopter crash, into a famous river in Sarawak, domiciled by huge crocodiles, is discussed in this paper. Search and recovery team as well as the combat divers from the Special Elite Troop Commando, known as VAT 69, were deployed to the scene to perform the underwater recovery to search for all the victims on board involving five Malaysians with a pilot of Philippines nationality. This paper highlights the limitations and challenges faced during the underwater search and recovery. All the bodies recovered were in moderate decomposition stage with crushed injuries and mutilated face and body. A simple and conventional fingerprint technique were used to record the fingerprint. The prints impressions were later photographed using a smartphone and transferred back to the RMP headquarters in Kuala Lumpur for fingerprint match by using WhatsApp Messenger, a phone application. All the first five victims were identified within an average of 10min. The last victim recovered was the pilot. For foreign nationals, the Immigration Department of Malaysia will record the prints of both index fingers only. The lifting of the fingerprint of the last victim was the most challenging in which only one index finger left that can be used for comparison. A few techniques were attempted using the black printer's ink, glass and tape techniques for the last victim. Subsequently, images of the prints impression were taken using the same smartphone with additional macro lens attached to it to enhance the resolution. The images were transferred to the RMP headquarters through WhatsApp Messenger. The prints were confirmed to be the pilot 20min later.
The VoIP services provide fertile ground for criminal activity, thus identifying the transmitting computer devices from recorded VoIP call may help the forensic investigator to reveal useful information. It also proves the authenticity of the call recording submitted to the court as evidence. This paper extended the previous study on the use of recorded VoIP call for blind source computer device identification. Although initial results were promising but theoretical reasoning for this is yet to be found. The study suggested computing entropy of mel-frequency cepstrum coefficients (entropy-MFCC) from near-silent segments as an intrinsic feature set that captures the device response function due to the tolerances in the electronic components of individual computer devices. By applying the supervised learning techniques of naïve Bayesian, linear logistic regression, neural networks and support vector machines to the entropy-MFCC features, state-of-the-art identification accuracy of near 99.9% has been achieved on different sets of computer devices for both call recording and microphone recording scenarios. Furthermore, unsupervised learning techniques, including simple k-means, expectation-maximization and density-based spatial clustering of applications with noise (DBSCAN) provided promising results for call recording dataset by assigning the majority of instances to their correct clusters.
The post-mortem resilience of facial creases was studied using donated bodies in order to establish the efficacy of crease analysis for identification of the dead. Creases were studied on normal (pre-embalmed) and bloated (embalmed) cadavers at the Centre for Anatomy and Human Identification (CAHID) to establish whether facial bloating would affect facial crease visibility. Embalming was chosen to simulate the effects produced by post-mortem bloating. The results suggested that creases are resilient and changes were only detected for creases located on the periphery of the face, particularly at areas where the skin is thick, such as at the cheeks. Two new creases not previously classified were identified; these creases were called the vertical superciliary arch line and the lateral nose crease. This research suggests that facial creases may be resilient enough after death to be utilised for human identification.
Video vision mixer based skull-photo superimposition is a popular method for identifying skulls retrieved from unidentified human remains. A report on the reliability of the superimposition method suggested increased failure rates of 17.3 to 32% to exclude and 15 to 20% to include skulls while using related and unrelated face photographs. Such raise in failures prompted an analysis of the methods employed for the research. The protocols adopted for assessing the reliability are seen to vary from those suggested by the practitioners in the field. The former include overlaying the skull- and face-images on the basis of morphology by relying on anthropometric landmarks on the front plane of the face-images and evaluating the goodness of match depending on mix-mode images; the latter consist of orienting the skull considering landmarks on both the eye and ear planes of the face- and skull-images and evaluating the match utilizing images seen in wipe-mode in addition to those in mix-mode. Superimposition of a skull with face-images of five living individuals in two sets of experiments, one following the procedure described for the research on reliability and the other applying the methods suggested by the practitioners has shown that overlaying the images on the basis of morphology depending on the landmarks on the front plane alone and assessing the match in mix-mode fails to exclude the skull. However, orienting the skull relying on the relationship between the anatomical landmarks on the skull- and face-images such as Whitnall's tubercle and exocanthus in the front (eye) plane and the porion and tragus in the rear (ear) plane as well as assessing the match using wipe-mode images enables excluding that skull while superimposing with the same set of face-images.