Numerous kinds of fingerprint powder such as aluminum flake have been successfully used for latent fingerprint detection in investigation. These fingerprint powders are hazardous to human health, especially for investigators, and sometimes seriously damage valuable furniture, electronic devices and precision instruments.
  In this paper, dogtooth violet starch, wheat flour, black tea, green tea, green laver were ground into fine particles less than 80 micrometers using a ball mill grinder. The shape of these fine powders were observed by scanning electron microscope (SEM), and their adherence to the skin ridge deposits was evaluated. Each powder detected a latent fingerprint 4 hours after imprinting on a glass plate, but they did not give good results for older or weaker fingerprints.
  Green laver, which is a marine plant of the green algae family, has a single cell layer structure. An improved green laver powder, produced by freeze-drying followed by grinding, contains a remarkable amount of thin plate particles. This characteristic feature of the improved green laver powder made it more adhesive, revealing even weak latent fingerprints 3 days after imprinting on a glass plate. The quality of lifted fingerprints was good enough to be used for fingerprint identification.

抄録全体を表示

  Behavioral case linkage is the method of identifying crimes committed by the same offender analyzing offender's behavior in crime scenes. Previous researches on behavioral case linkage have not provided enough information of analyzing procedure and predictive validity of quantitative case linkage in practice, especially in case of number of serial offences is more than four.
  In the present study, we illustrated the procedure of quantitative case linkage for practice, and investigated predictive validity of behavioral case linkage analysis, conducting a simulation experiment analyzing data from 90 serial rape cases of 18 offenders by hierarchical cluster analysis and non-metric multidimensional scaling (MDS).
  The results of the study indicated that 66% of the trials were perfectly or satisfactorily performed (average Cohen's kappa=0.78) in a simulation experiment of hierarchical cluster analysis. The results also indicated that MDS could effectively illustrate association between serial rapes by different offenders in most of trials.
  It is considered that quantitative case linkage based on behavior has enough predictive validity for helping crime investigations to link a number of serial crimes to one offender and reduce the cost of investigation of serial crimes.

抄録全体を表示

  Serial arsonists (N=125) were differentiated into four groups on the basis of their social independence and the degree of criminality, by using categorical principal components analysis (CATPCA). Results indicated the following. (1) Arsonists with high social independence and high criminality were aged 40 years or more in age, did not live with their parents, and tended to have criminal records for theft. (2) Arsonists with high social independence and low criminality were most frequently employed, high school graduates, married, and had no criminal record. Moreover, there were more female and mental patients in this group compared with other groups. (3) Arsonists with low social independence and high criminality were all male, compulsory education level (including high school dropouts), unmarried, not living with parent(s), and had no criminal record of theft. These arsonists tended to have relatively no criminal record of arson compared with other groups. (4) Arsonists with low social independence and low criminality were the youngest among four groups and were aged between 10 and 30 years, unmarried, lived with their parent(s), and had no criminal record. Results of log-linear analysis indicated that arsonists with high social independence tended to use a car or walk to the , whereas those with low social independence were inclined to use a bicycle. Moreover, arsonists with high criminality records tended to prepare the medium for arson in advance and drink alcohol before the offence, whereas arsonists with low criminality had a tendency to set fire to the same place repeatedly.

抄録全体を表示

This study aims to develop an intelligent security camera(or Public Security System) for detecting and reacting immediately to criminal acts, especially, a kidnapping case using automobile. We present a system for automated detection of criminal acts that uses the morphological information among detected objects in video sequences to classify the video scenes into kidnapping case or not. Some experimental results of scene recognition by using skeleton analysis are shown. Skeleton analysis is very efficient to represent a structure of the target body. We also consider some technical aspects of the intelligent security camera.

抄録全体を表示

  If the suspect left his/her fingerprints at a , depending on the types of crimes, i.e. whether it was a crime by the resident of the scene, crime by the person who had been in the scene, or crime by the person who discovered the crime, etc., the time when they were impressed often becomes a point of issue in the trial.
  Although some data on the change in the quality of fingerprints over time were already reported, they were based only on those that were over a short period of time.
  The change of the quality of fingerprints varies by individual differences, conditions when they were impressed, types of objects on which fingerprints were impressed, and conditions of places where fingerprints were kept.
  In this report, we conducted the experiment on the change of quality of fingerprints over a long period of time (seven years ago from now, and to be continued for another three years).
  The data of aging impressed fingerprints will contribute to scientific investigation.

抄録全体を表示

  At present, the footprint identification is conducted by the number of distinguishing characteristics point in the footprint. Life size footprints photos were usually used and magnified footprints photos were seldom used.
  If these characteristics parts were magnified and showed minute distinguishing marks more evidential values could be added to them.
  The method developed in this paper was to take pictures of magnified characteristic parts of both footprints retrieved at a and ones taken from shoes themselves in the same magnifying power, followed by enlarging photos from the negatives and comparing the minute distinguishing marks by superimposition.
  This method enables us to make magnified footprints photos of about 5 to 30 magnifications easily in a short time.
  By inspecting the magnified footprint photos, it was proved that one part which was simply evaluated as one characteristic in the present method actually had many distinguishing marks.

抄録全体を表示

  Shoeprints on soil or sand were lifted clearly with a quick setting resin. This resin, called Scandiquick, has been developed for metallographic embedding and consists of one powder (dibenzoyl peroxide, A) and one liquid (methyl methacrylate, B).
  Scandiquick was mixed in a volume ratio of two parts A and one part B using two measuring spoons, and both components were stirred thoroughly for approximately 30 seconds. Then the casting mixture was poured onto shoeprints. The resin set very quickly within 10 minutes and a clear impression appeared on it.
  The lifting method of shoeprints with Scandiquick has the following merits compared to that with plaster, which is the most popular for lifting shoeprints.
  1) This resin is especially effective for lifting shoeprints at the where water is not available.
  2) The set resin is light and tough, therefore, is easy to carry, handle and store.
  3) The impression of shoeprints can be photographed with transmitted light because the polymerized block has a slightly transparent appearance.

抄録全体を表示

This paper presents a computational framework for identity (initially about the culprit in a

) based on Barwise’s situation theory. Situations support information and can carry information about other situations. An utterance situation carries information about a described situation thanks to the constraints imposed by natural language. We are concerned with utterance situations in which identity judgments are made about the culprit in a , which is the corresponding described situation. The id-situation and along with various resource situations make up a case in the legal sense. We have developed OWL ontologies to provide concepts and principled vocabularies for encoding our scenarios in RDF, and we present an example of a SPARQL query of one of our encodings that spans situations. To follow how evidence supports hypotheses on the identity of the culprit in a , we use Dempster-Shafer theory. We tightly integrate it with our ontologies by having the representation of a case per our ontologies present a network containing situations and stitched together by objects; evidence "flows" along this network, diminishing and combining. We review the modifications of Dempster-Shafer theory required when one goes from a closed-world assumption to an open-world assumption. We review our plans regarding equational reasoning based on identities established in our id-cases, and we review the related issues regarding the meanings of URIs.

抄録全体を表示

  In the case of arson, the police often have difficulty in finding forensic evidence and eyewitnesses. If the police can identify the suspects based on their behaviours, it will be helpful for the criminal investigation.
  As the first step to fulfill the above purpose, this study focused on the behavioural characteristics of the Japanese repeat arsonists. It examined their behavioural consistency using the suspect retrieval support system developed by Adachi and his colleagues (1993, 1996, 1997). The system estimated behavioural similarities between a target sample and a database containing prior offenders and their offences, using the random choice probability method.
  The data comprised arsonists who had committed crimes between 1982 and 2005. Three hundred and nineteen repeat arsonists were the target sample in the above system. The offences in the database numbered 10,237, and these crimes were committed by 10,154 arsonists.
  Fifteen out of the 319 repeat arsonists were retrieved as a rank score of 1 among 10,154 offenders in the database, based on the similarity. The past records of 170 repeat arsonists were ranked within 10^th percentile in the database. Additionally, an examination of the offenders who had more than one offence recorded in the database revealed that the most recent crimes among the offences committed by the same offender were most similar to the target crimes.
  The consistency in the characteristics of each offence were compared by performing a chi-squared test between 81 offenders who showed the highest similarity and 81 offenders who showed the lowest similarity, differentiated by the suspect retrieval support system. The results showed that consistencies regarding the following behavioural characteristics showed highly significant differences between the two groups: the tools used and the way in which the fires were set, the objects burned, and the towns where the arsonists committed their crimes.

抄録全体を表示

  Offender profiling is one of the tools of decision making for criminal investigation. It is a set of techniques to infer characteristics of an unknown offender, such as sex, age bracket, lifestyle, psychological feature, previous crime, inhabited area, from the information which is left at the .
  In this article, we proposed a tool of decision-making for criminal investigation from the perspective of prediction of an uncertain event by the use of a Bayesian Network (BN). BN is a probability model that describes causal structure of events as chain networks of conditional probability, and is capable to predict the possibility of uncertain events.
  To examine the validity of the constructed model, firstly, we divided previous offenders’ information of the indoor-sex-offence cases into a training data (9,859 cases) and validation data (50 cases). Secondly, we constructed a model from the training data by means of K2 and MDL (minimum description length) as search-algorithm and information criteria, respectively. Finally, the validity of the model was examined by the validation data as virtual cases.
  According to the model, 21 target variables (16 behavioral variables, 2 vehicle variables and 3 victim variables) linked the explanatory variable (employment) directly, and most of these variables related to the employment. The results of the model validity showed that the accuracy of predicting the employment increased 10% higher when the age bracket could be estimated from the testimony of the victim.
  The results indicated that the BN model of the offender profiling would be able to provide valuable information for decision making for crime investigation. To predict characteristics of an unknown offender more accurately, it is crucial to select more appropriate information criteria and develop the search-algorithm, as well as to construct the database from more accurate information.

抄録全体を表示

  This paper describes an application of route simulation algorithms to decision making support system using digital map. especially the emergency deployment support system for Police.
  We proposed the adaptive route search algorithms in order to caluculate movable region from one poin in high precision and high speed as follows, (1) search process of candidate node with minimum traveling time, (2) restricted renewal process of accumulated cost of each node. And We develope the emergency deployment support prototype system for Police.
  The results of evaluation obtain that the proposed algorithms are about 4 seconds processing time by 3050R workstation, and the precision of 99% and calculation time of 60% compared with Dijkstra algorithm. So, the effectiveness of proposed algorithms are confirmed by this experiment.
  We have development a fast route simulation based on road conditions. This simulation calculates the area suspect could reach in a given time, allowing police commander deploy forces more efficiently than the customary circular area. In additions, it also locates the patrol-car closest to the and instruments it to rush there.

抄録全体を表示

  In a poisonous crime case, clarifying identity of the poisonous materials between actually used in the and the confiscated from the suspect's neighborhood is required.
  In this report, the examination of the discriminating capability of these compounds was conducted, using 13 As₂O₃ samples of known origins and manufacturing methods.
  By the observation using the scanning electron microscope, three morphological differences were recognized. The first one exhibited an euhedral crystal with octahedral shape, the second one had spherical shape and the third one was crystal showing significant cleavages. Furthermore, by energy-disperse X-ray analysis disclosed, only L_α,β and weak K_α,β peaks of arsenic in all samples, and the discrimination was impossible. Because of this reason, inductively coupled plasma-atomic emission spectrometry (ICP-AES) was adopted for the discrimination of the As₂O₃ samples by trace impurities. The compared heavy elements were barium (Ba), lead (Pb), antimony (Sb), selenium (Se) and tin (Sn) which were selected by a pre-qualified analysis of arsenous samples.
  As a result, significant difference was observed among these samples by comparison of the 5 elements, and their patterns were classified into three categories.
  It was considered that estimation of the origins and manufactured methods were possible from the morphological observation by scanning electron microscope and the comparison at trace impurities by ICP-AES.

抄録全体を表示

  This report is a case in which some bloodstains found in a were A₃B variant of the ABO blood group. A man was killed in a car and thrown into the sea. Five months later, several bloodstains were discovered in the car where he was killed. The reaction of those bloodstains by the absorption-elution test showed that A-antigenicity was much weaker than B-antigenicity. Therefore, it could not be decided whether the blood type of the bloodstain was AB or B. We investigated to get information about his blood type, because the blood group of the bloodstain were inexplicable. In the postmortem, his blood type and the characteristic of his teeth had been investigated to confirm him. It was estimated that his blood type was AB by the results of the absorption-elution test using decayed blood in his thoracic cavity.
  On the other hand, his ABO genotyping was judged to be AB type by gene analysis using DNA of the pulp of a tooth. Also, his blood type was AB, exarmined using the blood from the hospital. These information could not convince us, and we further investigated his blood type. As a result, we found that his blood group had been elucidated to be an A₃B type at a hospital that he was hospitalized at in the past. It was found that the cause of the indefinite blood type reaction was a variant.

抄録全体を表示

Small particles of gold foil detached from an indoor decoration might be important evidence to associate a suspect with a . We have investigated the application of elemental analysis using synchrotron radiation X-ray fluorescence spectrometry to discriminate small particles of gold foil. Eight kinds of gold foil samples collected in Japan were used in the experiments. As a result of synchrotron radiation X-ray fluorescence spectrometry, only two elements, gold and silver, were detected from all gold foil samples. The intensity ratios of AgK_α/AuL_α showed good correlation with the content ratios of Ag/Au. The variation of intensity ratio within a same sample was sufficiently small compared with those of different samples. Therefore the comparison of this intensity ratio can be an effective method to discriminate small particles originating from different types of gold foil.

抄録全体を表示

  In a recent murder case, many feathers were left at the and collected for analysis. It seemed they were likely left from the suspect's torn jacket. Goose and duck downs are commonly used for clothes and bedclothes, especially in high quality goods where goose feathers are used most often. Unfortunately, at the time of the murder there were few studies in Japan about down identification. This paper presents how to identify goose and duck downs by microscopy.
  Ten downs were removed at random from each stuffed bird from sixty-one species at the prefectural museum. Ten downs were removed at random from ten geese and ten ducks, respectively, at the prefectural farm. Therefore, the authentic sample set (family or species known) included ten downs each from eighty-one birds, representing sixty-three species. In addition, two hundred goose downs and two hundred duck downs were obtained from samples supplied by the Japan Spinners Inspecting Foundation in Tokyo.
  These down samples were examined microscopically with respect to eight morphological characteristics: full length, color, node shape, maximum node width, maximum node interval, node distribution, node density (number of nodes per mm) and pigment distribution. Morphological data from geese were compared with ducks and analyzed statistically using F-test.
  Duck and goose downs are identified primarily by their triangular nodes. In birds of the sixty-three species other than those from the duck and geese species, triangular nodes were found only in the Anatidae, Columbidae and Psittacidae families. Fortunately, it was quite simple to distinguish the families by the node distribution along the shaft of the barbules. For example, the Anatidae family has triangular nodes only toward the tip of the barbule, the Columbidae family has them mainly toward the base of the barbule, and the Psittacidae family has them uniformly distributed along the shaft of the barbule. Based on feather nodes, both goose and duck can be placed in the Anatidae family. Nevertheless, they can be distinguished. Goose has wider maximum node intervals than the duck, usually more than fifty-five micrometers. On the other hand, duck has higher node density than the goose, more than sixteen per mm. Statistical analysis using the F-test showed that the maximum node interval and node density were useful characteristics for distinguishing a goose from a duck down.

抄録全体を表示

  Soil can be one of the most important physical evidence in a criminal investigation which contributes to prove a linkage between the suspect and the . In this work, the validity of multiple instrumental analytical techniques for forensic soil identification was studied using 148 soil samples collected from the Tokyo metropolitan area. The samples were characterized by X-ray diffraction (XRD) analysis, X-ray fluorescence (XRF) analysis, magnetic susceptibility determination, scanning electron microscopy with an energy dispersive X-ray spectrometer (SEM-EDS) and high-energy synchrotron radiation X-ray fluorescence (SR-XRF) analysis. It was found that soil samples collected from railway stations and roadside dust showed higher magnetic susceptibilities (≧0.5×10^-4 m³ kg^-1 for 25% of the samples collected ) than those of soil samples collected from rivers, ponds, and lakes (<0.1×10^-4 m³ kg^-1 for 40% of the samples colleted). Microscopic examination of the soil samples revealed three kinds of characteristic spherules: i.e., transparent, white, and black particles. The SEM-EDS analyses of the spherule samples showed the following compositional characteristics. The transparent spherules exhibited two compositional types: Na-Ca-Si-O and Ba-Ti-Si-O types. On the other hand, black ones were either Fe-O or Si-Mg-O types and white ones were either Al-O or Sr-Al-O types. It was concluded that these six types of spherules were found at characteristic points and consequently, they can be used as forensic soil identification.

抄録全体を表示

  Soil can provide important information to criminal investigations as transfer evidence because many criminal cases take place under circumstances such that soil transfers to a criminal or victim. The variation in soils from place to place makes soil valuable evidence to prove linkage between a suspect and a . Soil is a complex mixture with a variety of mineralogical, chemical, biological, and physical properties. Considering such complexity, a variety of methods have been developed for forensic science purposes. Because minerals are an important component of soils, mineralogical examination is essential in forensic soil identification. Additionally, many other methods can be applied to raise the discriminating power, but not all kind of methods need to be used. What is important is that examiners select an appropriate combination of methods by considering the context of the soil samples. This report summarizes a wide range of reports on the analysis of soil components and of closely related materials such as plant fragments, pollen and spores, and diatoms, with emphasis on the importance of screening tests consisting of several simple techniques. The soil formation process involves parent materials, temperature, water condition, vegetation, time, and the chemical processes of solution, oxidation, reduction, and even human activities. The history of a soil's development as the results of such complex soil formation process is strongly reflected in soil color. The systematic observation of multiple soil colors is especially useful for screening.

抄録全体を表示

Many studies have indicated that most crimes occur close to an offender's home. However, recent studies demonstrated that substantial Journey-to-crime Distance (JtcD) variation was observed at the inter-offender level. This study aimed to examine whether the relationship between a central point calculated from a distribution of crime sites (Center of Minimum Distance: CMD) and each crime site and the

characteristics (students vs. other victims and indoor vs. outdoor settings) could explain the variation in JtcD at the inter- and intra-offender level. The data comprised 222 serial sexual offenders who committed seven or more offenses from 2003 to 2015 in Japan. The analyses of generalized linear mixed modeling showed that the CMD model constructed from the relationship between CMD and each crime site accounted for 47.9% of intra-offender variance in JtcD compared to the random intercept model. Additionally, when characteristics are included in the CMD model, the model accounted for 35.6% inter-offender variance in JtcD compared to the random intercept model. These findings indicated that the substantial variation in JtcD was explained by the relationship between CMD and each crime site and the characteristics, but the extent of this influence varied between offenders.

抄録全体を表示