To install StudyMoose App tap and then “Add to Home Screen”
Save to my list
Remove from my list
Cognitive abilities of human expertise modelled using computational methods offer several new possibilities for the forensic sciences. Typical areas include establishing a scientific basis for the expertise, providing tools for use by the forensic examiner, and providing an alternate opinion on a case. Organised criminality is a great concern for national/international security. The orchestration of complex crimes is increasingly dependent on knowledge distributed within law-enforcement agencies and scientific disciplines. Literature has already pointed out that forensic case data are amenable for use in an intelligence perspective if data and knowledge of specialised actors are appropriately organised, shared and processed.
The objective of the study is to discover a better and more extensible way of enhancing, expanding and empowering the capacity of Forensic Sciences Evaluation in Blood Spatter Analysis, not only with the assistance of current sciences and techniques, but also with future AI innovations.
Keywords - Forensic science, Computer science, Artificial intelligence, Law enforcement, Intelligent forensics, Artificial Intelligence, Feature Extraction, Region Props, Neural Network, Bloodstain Pattern Analysis, Cognitive Robotics, Pattern Recognition, Image Processing, Fuzzy Reasoning, Principal Component Analysis
The term "Artificial Intelligence" has been associated with several disciplines of human expertise.
Verified writer
Proficient in: Artificial Intelligence
4.9 (247)
“ Rhizman is absolutely amazing at what he does . I highly recommend him if you need an assignment done ”
Examples are computer-assisted vision, computational physical and analytical chemistry, etc. In the same context, knowledge and methods related to forensics can also be subjected to various AI methodologies.
The present set of forensic tools are not robust enough to analyse and correlate the results with a large amount of evidence. As a result, excessively time-consuming work are being handled by forensic experts.
Newer technology based approaches help the scientific community with methods to handle copious amounts of data which is not possible to handle through human hands, represent human expert knowledge in an efficient, structured way and for administering recognition and rational abilities in artificial machines.
Broadly, an artificial machine harnessing the power of artificial intelligence methods and algorithms will allow a forensic examiner or practitioner to:
? detect and enhance trace evidence for more exploratory investigation
? investigation and identification of evidence in an reproducible and objective manner
? examination methodology quality assessment
? reporting and standardising procedures of investigative practices
? probing for data from large dataset efficiently
? efficient visualisation of results of analysis
? serve as a great interpretation tool for results reporting and argumentation
? forecast unknown patterns or links to deduce new rules thus contributing to creation of more detailed knowledge bases
Blood is one of the most prevalent types of proof discovered in all violent crime scenes. Examination of bloodstain patterns has a coherent 150-year history and predates many contemporary forensic followers (Bevel, 2001). Currently, it is feasible to provide important data about the recognition and possible individualisation of an individual who has borne even the smallest quantity of blood through the science inquiry of blood at a scene of the crime. The analysis of bloodstain patterns can provide an understanding into the physical crime scene dynamic and in certain circumstances, for example in crimes when there are no eye-witnesses are available or for those that were present provide conflicting statements. [3]
Bloodstain pattern analysis (BPA) is defined as the scientific study of the static consequences resulting from dynamic blood shedding events [1]. BPA uses principles of biology (behaviour of blood), physics (cohesion, capillary action and velocity) and mathematics (geometry, distance, and angle) to assist investigators in answering questions such as: where did the blood come from?, what caused the wounds? and from what direction was the victim wounded?. A detailed study of bloodstain patterns at crime scenes often develops invaluable evidence. The distribution, size and shape of bloodstains on a victim, on a suspect, or on the walls, floors, ceilings, or on objects at the scene can help reconstruct these blood shedding events [1]. Bloodstain pattern analysis can also help one evaluate the credibility of statements provided by a witness, a victim, or a suspect [1] [2].
BPA plays an important role in conducting a reconstruction of the crime scene related to blood spatter events. By studying the distribution, size and shape of bloodstains, BPA enables worldwide investigation agencies in understanding the dynamics of a certain act of crime and evaluating the credibility of statements provided by a witness, a victim, or a suspect. However, in spite of its importance, this discipline is still mainly based on manual approaches, making the analysis of a crime scene time consuming, tedious and potentially not perfect. In Bloodstain pattern analysis (BPA), it is highly desirable to obtain non-invasively overall documentation of a crime scene, but also register in high resolution single evidence objects, like bloodstains. [4]
Currently, all the investigative activities are accomplished in a manual way resulting in an imprecise and slow reconstruction of facts. The imprecision and slowness are two aspects that are particularly important in the forensic area because they can negatively affect the life and wellbeing of people mistakenly involved in prosecutions for murder cases or, vice versa, they could not guarantee the just punishment for a person actually guilty a homicide. In order to overcome these drawbacks, re- cent researches in forensic science are attempting to design and develop enhanced tools for enabling investigators to perform an accurate and fast reconstruction of the crime scene related to bloody events and to infer the dynamics of violence acts. [2]
In cases of violent accidents of assault, homicide, abduction, and suicide, bloodstain pattern analysis (BPA) and its interpretation becomes a source of information of the occurrence of a crime, the possible sequence of events, and the position of individuals [1,2]. In many countries, BPA is a standard procedure at a crime scene where bloodstains are revealed. The gold standard of this forensic analysis is the presence of an expert directly at the crime scene to recognize the bloodstains and secure all the important information needed for the proper analysis of the creation mechanism. For a nonprofessional, it is extremely difficult to recognize and correctly describe bloody patterns, and to secure them such that they allow further analysis.
In addition, by using traditional CSI methods such as photography, reports, and sketches, it becomes almost impossible to secure information about the topography of a crime scene and all distances between potentially important objects present at a scene. The most common mistake with crime scene documentation are connected with the wrong dimensions for described objects or incorrect scales of the sketches. For proper analysis of bloodstains, it is crucial to combine all of the information about a single stain as well as the surrounding, as it is common to have stains that consist of many different blood spots on a large surface. For beginners and untrained BPA personnel, it is easy to focus only on part of the pattern and thus fail to consider the whole area. During the preparation of traditional photographic documentation, it becomes very easy to lose the overview, thus resulting in incorrect conclusions. All of the above, in combination with the absence of a BPA expert at the crime scene (which is common in many countries), might lead to a situation where, the documentation of the crime scene is incomplete or incorrectly prepared, thus preventing proper BPA. Thus, on one hand, it is almost infeasible to calculate the area of origin of blood based on the incomplete photographic documentation and report, but on the other hand it is very complicated and time consuming for the BPA expert to perform the analysis directly at the scene. To avoid common mistakes while documenting the scenes as well as secure evidenced in an effective and efficient manner that allows further analysis even by the untrained BPA personnel, we designed a platform to record crime scene with bloodstains in 3D and their subsequent evaluation. [4]
I. Blood Pattern Recognition & Analysis
Bloodstain Pattern Analysis (BPA) is used by forensic investigators to assess bloodstains left at crime scenes and gain information useful to be used for the reconstruction of incidents and the evaluation of the statements from witnesses and the crime participants. The consolidated method to determine the point from which the blood originated, named point of origin, is known as stringing technique [12]. The name of this method comes from the way in which forensic expert analyses the bloodstains.
In particular, once identified a spatter pattern, the analyst attaches elastic strings at the tip of the bloodstains' ellipse and extends them backward, or 180? opposite of their individual directions of travel. The two-dimensional point, named point of convergence, is where the strings intersect represents the two-dimensional geographic location of blood source. Determining the point of origin combines the two-dimensional point of convergence plus the angle of impact ? of the stains belonging to the pattern. Precisely, analysts pull elastic strings from the surface according to the angle ?. The angle of impact is the angle at which a blood droplet impacts a surface, measured with respect to an imaginary line perpendicular to that surface.
[3] present a proposal for a robotic framework to automate the BPA in all its steps. In particular, the robotic framework is composed of an Unmanned Aerial Vehicle (UAV) capable of navigating the crime scene, detecting bloodstains, computing the points of origin and preparing a technical report describing the bloody event.
By following this emergent trend, we propose to accomplish an automatic and fast BPA through the exploitation of forensic cognitive robots. In detail, the forensic cognitive robots are able (1) to navigate the post-bloody event scenario, (2) to detect the collection of bloodstains generated by the violent act, (3) to apply an automatic BPA pattern recognition engine to identify the collection of angles of impact, points and areas of convergence of the blows given by the attacker to the victim; and, finally, (4) to provide human investigators with a report about the reconstructed information.
Hence, the designed forensic robots can take the adjective cognitive since they do not exhibit only classical robotic functions such as navigation but more semantic capabilities in world such as context interpretation, decision-making and human-robot communication [5]. Precisely, the robot will navigate the crime scene by using a simple wall following logic together with a collision avoidance mechanism which employs only five off-the-shelf range sensors. As for the bloodstain recognition approach, the robot will exploit a visible wavelength hyper- spectral imaging based on the Soret ? band absorption in haemoglobin which allows to identify bloodstains in a non contact and non-destructive manner. Finally, the robot will perform the BPA by applying the Fuzzy C-Means clustering algorithm to the features extracted by above modules. The exploited robots are small Unmanned Aerial Vehicles (UAVs) capable of flying autonomously. This kind of robot allows to minimize the so-called scene contamination, i.e., the introduction of something to a scene that was not previously there, with respect to ground robot.
In BPA different 3D techniques may be applied to provide specialists a better insight into what could likely have occurred and enable an easier way to conclude about the possible sequence of events, relative position of individuals and objects. The overall 3D model of a scene is obtained via integration of laser scans registered from different positions. Some parts of a scene being particularly interesting are documented using midrange scanner, and the smallest details are added in the highest resolution as close-up scans. The scanning devices are controlled using developed software equipped with advanced algorithms for point cloud processing.
The most common technique is a laser scanning offering digitization of scanner's surrounding in few minutes and nowadays, it is becoming a standard in 3D crime scene documentation. However, resolution of laser scanners remains insufficient to analyse bloodstains with high accuracy. Considering this, a combination of different technique is highly required. To begin with, 3D scans may be complemented with traditional photographs. Commercial software such as Faro Scene provides tools to map photographs on 3D data.
Another widely used technique is photogrammetry which, as a low-cost and portable method, is especially attractive in daily forensic application. Fusion of the photogrammetric technique and laser scanning was applied in crime scene documentation and in BPA, where calculated trajectories and centres of origin were analysed and presented in a simplified CAD model, but placing additional markers in scene was also essential to obtain complete 3D model. In most complicated cases of bloodstains revealed on surface or multiple surfaces, which are not flat but rather have more complex geometry, both approaches with photographs mapping or photogrammetry fail with documentation of bloodstains for BPA.
AbstractCognitive abilities of human expertise modelled. (2019, Dec 02). Retrieved from https://studymoose.com/abstractcognitive-abilities-of-human-expertise-modelled-example-essay
👋 Hi! I’m your smart assistant Amy!
Don’t know where to start? Type your requirements and I’ll connect you to an academic expert within 3 minutes.
get help with your assignment
