Read Aloud the Text Content

This audio was created by Woord's Text to Speech service by content creators from all around the world.

Text Content or SSML code:

GOOD DAY TO EVERYONE, I THANK THE ORGANIZERS FOR THE OPPORTUNITY THEY GIVE ME TO PRESENT THIS WORK, WHICH WILL FOCUS ON ESTABLISHING A TECHNIQUE THAT ALLOWS US TO EVIDENCE THE RANGE OF VARIABILITY OF A HANDWRITTEN SIGNATURE, AND ESTABLISH ITS DEGREE OF VARIABILITY IN THESE ELEMENTS AND QUALITY TO BE USED WITHIN AN EXPERT STUDY OF IDENTIFICATION OF AUTHORSHIP OR GRAPHIC ORIGIN. MY NAME IS EDUARDO ACEVEDO, I AM FROM MEXICO, AND I CURRENTLY DEVELOP AS AN EXPERT IN ANALYSIS OF HANDWRITTEN SIGNATURES, AND I COLLABORATE AS A ASSISTANT IN THE ADMINISTRATION OF JUSTICE FOR THE JUDICIAL BRANCH OF MY PLACE OF MY RESIDENCE. The handwritten signature has always been a distinctive feature for the identification of a person over time. The variations that we can find in a writing or a signature generally correspond to the transformations that occur in the neural brain architecture, and these variations will be smaller or larger depending on the quality of their plasticity and the constant repetition of them. patterns, this is how the writer acquires his automation. In reality, what the expert in analysis of handwritten signatures does is compare patterns crystallized in the writer's brain and that have been laminated on a support. In this comparison we look for patterns that should be as similar as possible, that appear the same or are totally different. , the task is sometimes easy but other times it is difficult and can lead to errors in determining authorship, which is why it is necessary to incorporate tools into our profession that help avoid the errors that we can humanly make. For a long time, experts have claimed the ability to recognize variations in handwriting or handwritten signatures based on their “feeling” and “experience,” and I ask the following question: IS THERE CURRENTLY ANY WAY TO EVIDENCE IN A CLEAR AND OBJECTIVE WAY THE RANGE OF VARIABILITY OF A SET OF ELEMENTS THAT HAVE BEEN PROVIDED TO US AS COMPARISON ELEMENTS IN AN EXPERT STUDY? WITH THE RISK OF BEING WRONG, I CAN STATE THAT NO, AT LEAST AS REGARDS THE AMERICAN PART, BECAUSE I KNOW VERY LITTLE ABOUT WHAT IS CARRIED OUT IN THIS PART OF THE WORLD BUT I THINK THERE SHOULD NOT BE MUCH DIFFERENCE IN THIS PARTICULAR POINT. I know that many times our anchors and beliefs make us reflect on these new tools, but we must be flexible and take them into account if they make us more objective and improve the results we can obtain. This is precisely what we are presenting, new tools that can help minimize these errors, giving more reliability to our reports. Many studies now highlight concerns that the nature of the evidence and human error may call into question the reliability of many types of forensic examinations, including handwriting. The Office of Research and Forensic Sciences of The National Institute of Justice (NIJ) and Office of Special Programs of the National Institute of Standards and Technology (NIST) sponsored the work of an expert panel to encourage and enhance efforts to analyze the human factor focused on forensic analysis applications, reducing the risk of error and improving the practice of forensic examination of writings and signatures In this sense, the community of forensic experts in the analysis of writings and signatures in the United States and several countries have long introduced a series of characteristics that can be measured in a writing or signature, these being the ones that represent the greatest significance for our work the following DETERMINATION OF THE DEGREE OF COMPLEXITY CURVATURE OF STROKES LENGTH OF STROKES BIAS/TILT/ORIENTATION of stroke trajectories All these characteristics that ARE MEASURABLE, AND DEPENDING ON THE NUMBER OF SAMPLES under study, allow us to clearly and objectively establish the RANGE OF VARIABILITY of a writing or a signature, as pointed out by several authors, currently being the most Important in this development of proposals are well-known experts such as OSBORN, BRYAN FOUND, LINTON MOHAMED, MICHAEL CALIGIURY, HEIDI HARRALSON, in Latin America, LUIS GONZALO VELAZQUEZ POSADA, among others. From the above, we can then conclude that for forensic experts in analysis of writings and signatures, it is of utmost importance to MAKE MEASUREMENTS THAT ALLOW THEM TO REDUCE THE ERROR PRODUCED BY THE HUMAN FACTOR, and consequently have a MORE OBJECTIVE CONCLUSION, in line with a METHODOLOGY WITH MORE SCIENTIFIC BASES as recommended by the institutions that regulate forensic practice today. This work proposes the development of a technique using digital image processing using angular filtering in order to analyze the motor habits and developments of an individual that are reflected in the development of a signature and that these represent the automation process. that has been generated throughout a person's learning time, and if they are executed in a skillful, spontaneous and natural way, they represent a GRAPHIC PATTERN INDIVIDUAL, which in turn allows us to establish a RANGE OF VARIABILITY, depending on the quantity of authentic elements available. The basis of this work is the obtaining of characteristics that will be represented in a graph that will define the INDIVIDUAL GRAPHIC PATTERN model, and starting from this, be able to establish the RANGE OF VARIABILITY that is presented in each individual, and thus be able to evaluate the quality of the comparison elements, this as part of a PRE-ANALYSIS, taking advantage of the potential that the use of image processing tools with their already established algorithms gives us, to represent GRAPHIC PATTERNS of images in two dimensions using the FAST TRANSFORM DE FOURIER. The FOURIER TRANSFORM is a mathematical transformation that relates temporal or spatial representations and frequencies in an image. This method is based on Fourier spectrum analysis. For an image, structures with a preferred orientation generate a periodic pattern with an orientation in the Fourier transform of the image, compared to the direction of the objects in the input image. It is based on the technique proposed by the Chinese Academy of Sciences, Beijing, for the directional analysis of linear patterns in images. An image contains descriptive information about the object it represents. Keep in mind that under this relatively broad definition of IMAGE, there are many representations not perceptible to the human eye. In this proposal, through image processing, it is proposed to establish as specific characteristics the ORIENTATION OF STROKES and the ACCUMULATED DISTRIBUTION OF STROKES (OT-DAT)©™®, that is, it is proposed to create one of those representations that are not perceptible to the human eye. . The result of this information processing is a histogram that represents the INDIVIDUAL GRAPHIC PATTERN (PGI) The proposed technique is used to infer the preferred orientation of the structures present in the study image. Calculates a histogram indicating the number of structures in a given direction. Images with completely isotropic content are expected to give a flat histogram, while images in which there is a preferred orientation are expected to give a histogram with a ridge at that orientation. In a simplified way we can establish that the technique used to obtain the INDIVIDUAL GRAPHIC PATTERN and the study of the dominant orientation is made up of the following steps. a. Document analysis: Try to analyze the structure of a document and understand the information contained in its components. b. Acquisition: Refers to the process of converting a document into a representation suitable for processing by the computer. Such acquisition and conversion can be carried out by a scanner, photographic or video camera, etc. and the result may vary depending on the digitization process used and the encoding method c. Normalization: Try to adjust the shape, size and position of the character with the aim of minimizing the variation between images of the same class. d. Binarization: Converts the received image into a binary image, thus separating the background from the objects to be analyzed. and. Processing: It is the use of the Fourir transform algorithm from which the histogram of orientations of each analyzed image will be obtained. F. Comparison of extracted features with the intended model In the images that we are observing we can see how a graph has been generated for each element that we wish to study, and in which we can see how the specific automatisms of this person are represented. Also, with the information obtained we can obtain the range of variability that each individual has, thus having a clear reference of differentiation between possible authors of handwritten signatures. This tool can also help us understand the natural behavior of people over time, being able to detect important modifications and relate them to external causes. This is the general proposal, hoping it has been of interest to you, It is important to comment that it is currently being applied in real cases where reports are generated, graphs are interpreted and more objective conclusions are reached, taking into account the recommendations that forensic analysis of handwritten signatures be supported by more scientific bases. Thank you for your attention