Morphometrics-based method for identifying species of flies by using posterior spiracles
The morphometric-based method for fly species identification using posterior spiracles of third instar larvae addresses the complexity of traditional methods by providing a standardized and efficient system for species identification, achieving high accuracy and cost-effectiveness.
Patent Information
- Application Number
- PCT/KR2024/001108
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-11-16
- Filing Date
- 2024-01-24
- Publication Date
- 2024-11-21
AI Technical Summary
Morphological species identification of fly larvae is challenging due to its high complexity and limited applicability, especially in forensic entomology, where traditional methods require larvae to mature into adults and DNA barcoding is time-consuming and costly.
A morphometric-based method using the posterior spiracles of third instar larvae, which involves image collection, data processing to calculate reference values, and classification using Error rate-based morphometrics (EBM) to identify fly species by measuring angles and lengths of slit pairs, providing a standardized and efficient species identification system.
The method achieves high accuracy in identifying fly species, particularly for the Calliphoridae family, with low correlation to growth stage, enabling rapid and cost-effective species identification without requiring additional preprocessing or narrowing the species range.
Smart Images

Figure KR2024001108_21112024_PF_FP_ABST
Abstract
Description
A method for species identification of flies based on morphometric analysis using epigrams
[0001] The present invention relates to a method for identifying fly species using morphological characteristics of the posterior spiracles of third instar larvae.
[0002] Forensic entomology is the study of using insect evidence to resolve legal issues. In forensic entomology, estimating the postmortem interval using insect larvae, particularly saccharomyces flies, is the most common method of estimating the postmortem interval in dead bodies. Therefore, saccharomyces fly larvae are arguably the most important forensic evidence in forensic entomology. Among saccharomyces flies, the family Blackfly is the first to approach and lay eggs on dead bodies, making saccharomyces fly larvae essential for forensic entomology practice. However, morphological identification of larvae is highly specialized and notoriously difficult. Therefore, the traditional forensic entomological method involved rearing larvae to adulthood and observing the adult morphology to confirm the species. DNA barcoding is a widely used alternative to morphological identification. DNA barcoding can identify many species using standardized methods such as polymerase chain reaction (PCR) and base sequence analysis, but it is time-consuming and expensive, and interpretation of the results requires knowledge of molecular evolution, which presents a barrier to entry.
[0003] Despite these challenges, morphological species identification offers the absolute advantage of being cost-effective and rapid. Indeed, studies exist that attempt to overcome the limitations of conventional morphological species identification. These studies utilize identical structures found across all species and employ statistical methods to identify multiple species using the same structures. A representative example is a study that used landmarks-based geometric morphometrics on adult fly wings, but this method is limited to adults. Another study utilized the cephalopharyngeal skeleton in larvae for landmarks-based geometric morphometrics, but this method requires preprocessing. While these studies utilized structures found across all species, they cannot be said to completely overcome the limitations of conventional morphological species identification due to their limited range of applicable species or the need for preprocessing.
[0004] In the present invention, we utilized the posterior pronotum, a structure common to all maggots, to develop a morphological species identification method for maggots that is applicable to multiple species, simple, and unified. Although many studies have mentioned the significance of the posterior pronotum as an indicator of species identification, a standardized method for analyzing species using the posterior pronotum of larvae has not yet been developed. In this invention, we propose a new indicator that can standardize the morphological characteristics of the posterior pronotum of third-instar larvae of flies, and we propose a method for identifying fly species of third-instar larvae using this indicator. In this invention, the validity of the fly species identification method was verified using larvae of nine species of Calliphoridae, which showed the highest occurrence rate in fatal cases in Korea.
[0005] [Prior Art Literature]
[0006] [Patent Document]
[0007] Korean Patent Publication No. 10-2099457 (Published on April 10, 2020)
[0008] The present invention relates to a system for identifying fly species using the morphological characteristics of third-instar larvae, and by utilizing three pairs of slits, which are morphological structures that all third-instar larvae of flies have, an index capable of standardizing morphological characteristics is discovered, and a new classification method called EBM (Error rate-based morphometrics) is devised based on the error rate to develop a new method for identifying fly species, and this system is applied to Calliphora lata (la), Calliphora vicina (vi), Triceratopyga calliphoroides (cl), Lucilia illustris (il), Lucilia caesar (ca), Lucilia sericata (se), Chrysomya megacephala (mg), Chrysomya pinguis (pi), and other fly species belonging to the Calliphoridae family. By applying it to the black goldfly (Phormia regina)(re), the validity of fly species identification is statistically verified, thereby providing a new method for fly species identification.
[0009] The present invention relates to a system for identifying flies using morphological characteristics of third instar larvae, comprising: an image collection unit for photographing the posterior spiracle of third instar larvae from a fly population and collecting images of the posterior spiracle; a data collection unit for extracting only posterior spiracle images of a standard fly population whose scientific name has been identified from the fly population, and measuring morphological characteristics from the posterior spiracle images of the extracted standard fly population to collect morphological data; a data processing unit for calculating reference values of the morphological data, including an average value, a standard deviation value, a minimum value, a maximum value, and a median value statistically calculated from the collected morphological data; a classification reference model construction unit for constructing a first classification reference model and a second classification reference model from the calculated reference values and classifying the standard flies according to the scientific names of the flies; a measurement collection unit for extracting only posterior spiracle images of an unknown fly population whose scientific name has not been identified from the fly population, and measuring morphological characteristics from the posterior spiracle images of the extracted unknown fly population to collect measurement values; And, a system for identifying a fly is provided, including an analysis unit for identifying a fly by comparing the reference values provided from the first classification reference model and the second classification reference model with the measured values.
[0010] In addition, the present invention provides a method for identifying a fly using a fly identification system utilizing morphological characteristics of third-instar larvae, comprising the steps of: collecting an image of the posterior pronotum of an unknown third-instar larva that has not been identified; identifying three pairs of slits symmetrical on both sides from the collected image, and recognizing them as a first slit pair, a second slit pair, and a third slit pair starting from the slit pair located most upward from the center of symmetry; measuring an angle (°) formed by the three pairs of slits to measure an angle (°) value ABS_1 formed by the first slit pair, an angle (°) value ABS_2 formed by the second slit pair, and an angle (°) value ABS_3 formed by the third slit pair; measuring the length of each of the second slit pairs to measure an average length (mm) value LBS (Length BiSlit); A method for identifying a fly is provided by a fly identification system, including: a step of measuring DBS (Distance Between Slit), which is a value obtained by dividing the shortest distance (mm) between the lower points of the second slit by the LBS; and a step of identifying a fly by comparing the measured values of ABS_1, ABS_2, ABS_3, LBS, and DBS with first and second classification reference models constructed according to the scientific name of the fly in the fly identification system.
[0011] According to the present invention, a method is provided for classifying Calliphora lata (la), Calliphora vicina (vi), Triceratopyga calliphoroides (cl), Lucilia illustris (il), Lucilia caesar (ca), Lucilia sericata (se), Chrysomya megacephala (mg), Chrysomya pinguis (pi), and Phormia regina (re) belonging to the Calliphoridae family by selecting an appropriate classification index that will serve as a standard for species identification by analyzing the morphological characteristics of the posterior phylum of the third instar larval, and the significance thereof is statistically analyzed to demonstrate that it exhibits high accuracy, and the classification index for species identification is provided for the larval By confirming that the degree of growth (body length) is not related to the species identification of the Calliphoridae flies, a new method of species identification of the black fly family (Calliphoridae) can be provided through the morphological image of the latter.
[0012] Figure 1 is a photograph of the posterior larva of a third instar fly belonging to the Calliphoridae family.
[0013] Figure 2 shows the results of measuring the classification indices of nine species of the black fly family (Calliphoridae) and creating boxplots corresponding to individual classification indices for each species. Figure 2a is for ABS_1, Figure 2b is for ABS_2, Figure 2c is for ABS_3, Figure 2d is for DBS, and Figure 2e is for LBS.
[0014] Figure 3 shows the results of applying the species identification method of a Calliphoridae fly according to the present invention to an unknown sample, and measuring five classification indices (ABS_1, ABS_2, ABS_3, LBS, and DBS) in the late third instar larval scutes of an unknown fly sample.
[0015] Figure 4a shows the results of a heatmap analysis of the correlation between three classification indices (ABS_2, DBS, and LBS) of the black goldfly (Phormia regina).
[0016] Figure 4b is the result of creating a scatter plot of the distribution pattern of ABS_2 for the body length of the black goldfly (Phormia regina).
[0017] Figure 4c is the result of a scatter plot of the distribution pattern of DBS with respect to the body length of the black goldfly (Phormia regina).
[0018] Figure 4d is the result of a scatter plot of the distribution pattern of LBS for the body length of the black goldfly (Phormia regina).
[0019] The terms used in this specification have been selected from widely used, current terms, taking into account the functions of the present invention. However, these terms may vary depending on the intentions of those skilled in the art, precedents, the emergence of new technologies, etc. Furthermore, in certain cases, terms may be arbitrarily selected by the applicant, and in such cases, their meanings will be described in detail in the relevant description of the invention. Therefore, the terms used in this invention should not be defined simply as names, but rather based on their inherent meanings and the overall content of the present invention.
[0020] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and shall not be interpreted in an idealized or overly formal sense unless explicitly defined herein.
[0021] Hereinafter, the present invention will be described in more detail.
[0022] In the present invention, a fly identification system using the morphological characteristics of third-instar larvae was developed, and new indices called ABS, DBS, and LBS were discovered by utilizing three pairs of slits as morphological characteristics of flies. ABS stands for Angle between Slits and refers to the angle formed by both pairs of slits. DBS stands for Distance between Slits and refers to the ratio of the length of the second slit to the distance between the two slits. LBS stands for Length BiSlit and refers to the average length of the second slit. All indices are measured based on the center of the distal end of the slit, and have the advantage of being applicable to Calliphoridae, Sarcophagidae, and Muscidae regardless of the slit shape. Based on the error rate of the three indices, a new fly species classification method called Error rate-based morphometrics (EBM) was developed. The validity of the fly species classification method according to the present invention was verified by applying it to nine species of the black fly family (Calliphoridae), and it was verified through a large number of black goldfly (Phormia regina) individuals that the newly discovered indicator is a morphologically valid value regardless of the growth stage of the fly larvae.
[0023] The present invention relates to a system for identifying flies using morphological characteristics of third instar larvae, comprising: an image collection unit for photographing the posterior spiracle of third instar larvae from a fly population and collecting images of the posterior spiracle; a data collection unit for extracting only posterior spiracle images of a standard fly population whose scientific name has been identified from the fly population, and measuring morphological characteristics from the posterior spiracle images of the extracted standard fly population to collect morphological data; a data processing unit for calculating reference values of the morphological data, including an average value, a standard deviation value, a minimum value, a maximum value, and a median value statistically calculated from the collected morphological data; a classification reference model construction unit for constructing a first classification reference model and a second classification reference model from the calculated reference values and classifying the standard flies according to the scientific names of the flies; a measurement collection unit for extracting only posterior spiracle images of an unknown fly population whose scientific name has not been identified from the fly population, and measuring morphological characteristics from the posterior spiracle images of the extracted unknown fly population to collect measurement values; And, a system for identifying a fly is provided, including an analysis unit for identifying a fly by comparing the reference values provided from the first classification reference model and the second classification reference model with the measured values.
[0024] The above morphological features include ABS_1, which is an angle (°) formed by the first pair of slits (US, Upper slits) located at the top among the three pairs of slits symmetrical on both sides of the posterior pronotum in a photograph of the posterior pronotum in which the mouthparts of the larva are arranged downward and the abdomen upward; ABS_2, which is an angle (°) formed by the second pair of slits (MS, Middle Slits) located in the middle among the three pairs of slits symmetrical on both sides of the posterior pronotum; ABS_3, which is an angle (°) formed by the third pair of slits (LS, Lower Slits) located at the bottom among the three pairs of slits symmetrical on both sides of the posterior pronotum; LBS (Length BiSlit), which is an average length (mm) of the second slits; and DBS (Distance Between Slit), which is a value obtained by dividing the shortest distance (mm) between the lower points of the second slits symmetrical on both sides of the posterior pronotum by the LBS.
[0025] The first classification criterion model is composed of ranges expressed as minimum and maximum values of the morphological data, and preferably, the first classification criterion model is the minimum and maximum values of LBS (Length BiSlit), which is the average length (mm) of the second slit. The second classification criterion model is composed of median values of the morphological data, and preferably, the second classification criterion model is composed of median values of ABS_1; ABS_2; ABS_3; LBS (Length BiSlit); and DBS (Distance Between Slit). The first classification criterion model and the second classification criterion model are determined based on morphological data collected from a standard fly population whose scientific name has been identified, and the classification criterion model varies depending on the fly species.
[0026] In addition, the present invention provides a method for identifying a fly using a fly identification system utilizing morphological characteristics of third-instar larvae, comprising the steps of: collecting an image of the posterior pronotum of an unknown third-instar larva that has not been identified; identifying three pairs of slits symmetrical on both sides from the collected image, and recognizing them as a first slit pair, a second slit pair, and a third slit pair starting from the slit pair located most upward from the center of symmetry; measuring an angle (°) formed by the three pairs of slits to measure an angle (°) value ABS_1 formed by the first slit pair, an angle (°) value ABS_2 formed by the second slit pair, and an angle (°) value ABS_3 formed by the third slit pair; measuring the length of each of the second slit pairs to measure an average length (mm) value LBS (Length BiSlit); A method for identifying a fly is provided by a fly identification system, including: a step of measuring DBS (Distance Between Slit), which is a value obtained by dividing the shortest distance (mm) between the lower points of the second slit by the LBS; and a step of identifying a fly by comparing the measured values of ABS_1, ABS_2, ABS_3, LBS, and DBS with first and second classification reference models constructed according to the scientific name of the fly in the fly identification system.
[0027] The step of identifying the fly includes: comparing the LBS measurement value with the first classification reference models to select a scientific name of a fly corresponding to the first classification reference model including the LBS measurement value as a classification reference species; comparing the reference values of the second classification reference models of the selected classification reference species with the measurement values of ABS_1, ABS_2, ABS_3, LBS, and DBS to calculate an absolute error rate; and identifying the unknown fly as a classification reference species in which the sum of the calculated absolute error rates is the lowest.
[0028] In the step of selecting the above classification criterion species, if only one first classification criterion model is selected, the unknown fly is identified by the scientific name of the standard fly corresponding to the selected first classification criterion model.
[0029] The above absolute error rate is a value calculated by substituting the reference value of the second classification criterion model and the measured value into [Mathematical Formula 1] below.
[0030] [Mathematical Formula 1]
[0031]
[0032] In addition, the inventors of the present invention obtained 151 samples of third instar larvae of Calliphora lata (la), Calliphora vicina (vi), Triceratopyga calliphoroides (cl), Lucilia illustris (il), Lucilia caesar (ca), Lucilia sericata (se), Chrysomya megacephala (mg), Chrysomya pinguis (pi), and Phormia regina (re) belonging to the Calliphoridae family as a method for species identification of flies in the Calliphoridae family, and analyzed the morphological characteristics of the posterior phylum of the third instar larval, and used the above morphological characteristics to identify the species of the Calliphoridae flies. A method for species identification of Calliphoridae flies is provided by designing a first classification criterion model and a second classification criterion model.
[0033] The above morphological features are composed of ABS_1, ABS_2, ABS_3, DBS, and LBS, wherein the ABS (angle between slits) refers to the angle (°) formed by bilateral symmetrical slit pairs of the posterior spiracle, the ABS_1 refers to the angle (°) formed by the first slit pair located at the top among the symmetrical slit pairs, the ABS_2 refers to the angle (°) formed by the second slit pair located in the middle among the symmetrical slit pairs, and the ABS_3 refers to the angle (°) formed by the third slit pair located at the bottom among the symmetrical slit pairs. The DBS (Distance Between Slits) refers to the ratio (distance / length) of the shortest distance (mm) between the second slit pairs and the average length (mm) of the second slits, and the LBS (Length Bislit) refers to the average length (mm) of the second slits.
[0034] In the present invention, among the collected morphological characteristics, the LBS index was statistically analyzed to confirm that the standard deviation range of the classification index for distinguishing the nine species of the Calliphoridae family was the smallest.
[0035] A method for identifying a species of a Calliphoridae fly according to one embodiment of the present invention comprises a first classification step using a first classification reference model and a second classification step of identifying the fly species by comparing the species selected through the first classification step with the first classification reference model.
[0036] Specifically, in one embodiment of the present invention, a method for identifying a species of a Calliphoridae fly using morphological characteristics of a third instar larva comprises the steps of: (1) preparing an image by photographing a posterior spiracle in a third instar larval of a fly to be identified; (2) confirming three pairs of slits symmetrical on both sides of the posterior spiracle in the image, and recognizing them as a first slit pair, a second slit pair, and a third slit pair starting from the slit pair located most upward from the center of symmetry; (3) measuring the length of each of the second slit pairs to measure an average length (mm) value, LBS (Length BiSlit); And (4) a step of selecting a classification reference species corresponding to an LBS reference range including the measured value of the LBS by comparing with the first classification reference model of the Calliphoridae fly shown in Table 1 below; a method for identifying a species of a Calliphoridae fly is provided.
[0037]
[0038] Specifically, in the first classification step, the range of LBS (mm) is used as a classification index, and the Calliphoridae fly can be identified by determining whether the LBS measurement value measured from the sample of the fly to be identified falls within the LBS standard range corresponding to each classification standard species. If the LBS measurement value exceeds 0.22 mm or is less than or equal to 0.26 mm, the large black fly (Calliphora lata) (la) is selected as the classification standard species. If the LBS measurement value exceeds 0.1475 mm or is less than or equal to 0.19 mm, the red-cheeked black fly (Calliphora vicina) (vi) is selected as the classification standard species. If the LBS measurement value exceeds 0.12 mm or is less than or equal to 0.1475 mm, the two-tailed black fly (Triceratopyga calliphoroides) (cl) is selected as the classification standard species. If the above LBS measurement value exceeds 0.17 mm or is 0.20 mm or less, Lucilia illustris (il) is selected as the classification reference species. If the above LBS measurement value exceeds 0.18 mm or is 0.22 mm or less, Lucilia caesar (ca) is selected as the classification reference species. If the above LBS measurement value exceeds 0.17 mm or is 0.21 mm or less, Lucilia sericata (se) is selected as the classification reference species. If the above LBS measurement value exceeds 0.28 mm or is 0.37 mm or less, Chrysomya megacephala (mg) is selected as the classification reference species. If the above LBS measurement value exceeds 0.27 mm or is 0.36 mm or less, Chrysomya pinguis (pi) is selected as the classification reference species. If the above LBS measurement value is over 0.23 mm or less than 0.27 mm, the black goldfly (Phormia regina)(re) is selected as the classification standard species.
[0039] In the first classification step, only one or at most four classification criteria species are selected based on the measured value of LBS. If only one classification criteria species is selected, the flies in the sample can be identified with the selected classification criteria species. If two or more classification criteria species are selected, a second classification step is performed. Specifically, if two or more classification criteria species are selected, (4) a step of measuring ABS_1, which is an angle (°) value formed by the first slit pair, ABS_2, which is an angle (°) value formed by the second slit pair, ABS_3, which is an angle (°) value formed by the third slit pair, and DBS (Distance Between Slit), which is a value obtained by dividing the shortest distance (mm) between the lower points of the second slits by the LBS; (5) a step of calculating an absolute error rate by comparing the reference values provided in the second classification reference model of the black fly (Calliphoridae) shown in Table 2 below with the measured values of ABS_1, ABS_2, ABS_3, LBS, and DBS; and (6) a step of identifying the fly as the classification reference species in which the sum of the calculated absolute error rates is the lowest.
[0040]
[0041] [Mathematical Formula 1]
[0042]
[0043] Specifically, if the large black fly (Calliphora lata)(la) is selected as the classification reference species in the first classification step, the reference value of ABS_1 is 89.0°, the reference value of ABS_2 is 129.6°, the reference value of ABS_3 is 164.5°, the reference value of DBS is 1.291, and the reference value of LBS is 0.2415 mm. If the red-cheeked black fly (Calliphora vicina)(vi) is selected as the classification reference species in the first classification step, the reference value of ABS_1 is 105.9°, the reference value of ABS_2 is 128.2°, the reference value of ABS_3 is 161.9°, the reference value of DBS is 2.354, and the reference value of LBS is 0.1687 mm. In the first classification step, if the two-tailed black fly (Triceratopyga calliphoroides)(cl) is selected as the reference species, the reference value of ABS_1 is 102.6°, the reference value of ABS_2 is 126.6°, the reference value of ABS_3 is 158.6°, the reference value of DBS is 2.457, and the reference value of LBS is 0.1333 mm. If the green-green gold fly (Lucilia illustris)(il) is selected as the reference species in the first classification step, the reference value of ABS_1 is 78.6°, the reference value of ABS_2 is 109.7°, the reference value of ABS_3 is 146.8°, the reference value of DBS is 1.454, and the reference value of LBS is 0.1884 mm. In the first classification step, if Lucilia caesar (ca) is selected as the reference species, the reference value of ABS_1 is 72.3°, the reference value of ABS_2 is 105.9°, the reference value of ABS_3 is 146.1°, the reference value of DBS is 1.226, and the reference value of LBS is 0.2046 mm. If Lucilia sericata (se) is selected as the reference species in the first classification step, the reference value of ABS_1 is 76.0°, the reference value of ABS_2 is 95.7°, the reference value of ABS_3 is 135.3°, the reference value of DBS is 1.975, and the reference value of LBS is 0.1919 mm.In the first classification step, if Chrysomya megacephala (mg) is selected as the reference species, the reference values for ABS_1 are 36.2°, ABS_2 is 67.0°, ABS_3 is 103.9°, DBS is 1.234, and LBS is 0.3165 mm. If Chrysomya pinguis (pi) is selected as the reference species in the first classification step, the reference values for ABS_1 are 47.5°, ABS_2 is 77.9°, ABS_3 is 115.2°, DBS is 1.161, and LBS is 0.3116 mm. In the first classification step, if the black goldfly (Phormia regina)(re) is selected as the classification reference species, the reference value for ABS_1 is 43.9°, the reference value for ABS_2 is 78.9°, the reference value for ABS_3 is 119.7°, the reference value for DBS is 1.550, and the reference value for LBS is 0.2470 mm.
[0044] Hereinafter, to aid understanding of the present invention, experimental examples and examples will be described in detail. However, the following experimental examples and examples are merely illustrative of the present invention and are not intended to limit the scope of the present invention. The experimental examples and examples of the present invention are provided to more fully explain the present invention to those of average skill in the art.
[0045]
[0046] <Experimental Example> Experimental Materials and Methods
[0047] The following experimental examples are intended to provide experimental examples commonly applied to each embodiment according to the present invention.
[0048]
[0049] 1. Sample collection
[0050] Sampling was conducted on third-instar larvae of the Calliphoridae family, which are major voracious flies found in Korea and have a high occurrence rate in real-world events. To exclude artificial influences that may occur in laboratory environments, samples were collected only from real-world events and pig carrion experiments. Third-instar larvae of nine species of Calliphoridae with a high occurrence rate in Korea were collected: Calliphora lata (la), Calliphora vicina (vi), Triceratopyga calliphoroides (cl), Lucilia illustris (il), Lucilia caesar (ca), Lucilia sericata (se), Chrysomya megacephala (mg), Chrysomya pinguis (pi), and Phormia regina (re). The above samples were collected from actual death cases in the Korea University Autopsy Laboratory (KU), Seoul, Incheon, Paju, Uijeongbu, and Busan, and from pig decomposition experiments conducted at the Korea University Deokso Farm (KU) and the Asan Police Training Institute (KPIA). The samples were boiled in hot water (≥80°C) for at least 30 seconds and then stored in 70% ethanol.
[0051]
[0052] 2. Securing basic data for classification indicators
[0053] Among the collected samples, only mature third-instar larvae were selected as samples, excluding early-stage third-instar larvae, damaged individuals, and contaminated individuals, and images of the anal division (Fig. 1) were taken. Images were collected using a SZX10 stereomicroscope (Olympus, Tokyo, Japan) and an ECMOS05300KPA digital camera (Sony, Konan, Japan), and the anal division was taken horizontally. Data were used only when all individual classification indices (ABS_1, ABS_2, ABS_3, DBS, and LBS) could be measured in the images taken for each sample. Ultimately, 151 individuals from 24 incidents were used as samples, and data were collected by measuring the classification indices (ABS_1, ABS_2, ABS_3, DBS, and LBS) for each individual. The species and classification of the finally selected samples are shown in Table 3 below. Individual taxonomic indices were measured using ToupView (ToupTek Photonics, Hangzhou, China). Species identification was performed using COI barcoding using the Exgene Tissue SV mini kit (GeneAll, Seoul, Korea).
[0054] SubfamilySpeciesNumber of samplesCollection sitesCalliphorinaeCalliphora lata(la)21Incheon, BusanCalliphora vicina(vi)18KU, BusanTriceratopyga calliphoroides(cl)14KULuciliinaeLucilia illustris(il)25KULucilia caesar(ca)13KULucilia sericata(se)13KU, SeoulChrysomyinaeChrysomya megacephala(mg)17Incheon, PajuChrysomya pinguis(pi)18Incheon, KPIAPhormia regina(re)12KU, Uijeongbu
[0055] *KU: Korea University Autopsy Room & Korea University Deokso Farm
[0056] *KPIA: Korean Police Investigation Academy
[0057] Basic statistical data for each species (Tables 4 to 8) were obtained for each indicator, including the number of individuals (obs), mean, standard deviation (sd), minimum (min), maximum (max), and median (median), using the statistical program R (ver. 4.2.3).
[0058] ABS_1(°)laviclicasemgpireobs211814251313171812mean91.6106.4102.377.472.874.938.047.843.6sd10.311.19.45.36.97.25.85.77.2mi n73.190.986.066.060.164.331.836.734.2max109.4131.2115.286.386 .384.951.655.856.3median89.0105.9102.678.572.376.036.247.543.9
[0059]
[0060] ABS_2(°)laviclilcasemgpireobs211814251313171812mean127.3130.3125.3108.8106.397.168.376.678.0sd9.29.19.05.45.47.68.37.06.2min106.7117.7102.597.398.081.957.864.569.0max143.8148.7139.1118.6117.0107.786.787.285.5median129.6128.2126.6109.5105.995.767.077.978.9
[0061]
[0062] ABS_3(°)laviclilcasemgpireobs211814251313171812mean164.4161.8157.8146.0146.5132.9105.1115.1118.7sd5.610.89.67.03.17.08.07.96.6min150.8142.1135.0132.4140.3119.193.6100.9106.9max173.7186.6170.2165.5152.0142.5122.9127.9128.6median164.5161.9158.6146.4146.1135.3103.9115.2119.7
[0063]
[0064] DBSlaviclilcasemgpireobs211814251313171812mean1.3022.4322.4641.4531.2171.9761.2611.1731.519sd0.0870.2610.2280.1190.1020.1100.1080.0860.128min1.1002.0372.0961.2241.0561.7501.0901.0291.249max1.4353.1302.8601.6771.3552.1431.4441.3511.756median1.2912.3542.4571.4571.2261.9751.2341.1611.550
[0065]
[0066] LBS(mm)laviclicasemgpireobs211814251313171812mean0.24080.16890.13360.18760.20140.1 9360.31960.31400.2463sd0.00810.01190.00650.00710.00850.00830.02120.01940.0090min0.22 110.14960.12580.17230.18770.17740.28330.27140.2347max0.25590.18810.14560.19790.2134 0.20990.36090.35060.2626median0.24150.16870.13330.18880.20460.19190.31650.31160.2470
[0067]
[0068] 3. Accuracy verification
[0069] In order to verify the accuracy of the method for identifying black flies according to the present invention, verification was performed in the following three-step order using the statistical program R (ver. 4.2.3).
[0070] Verification Step 1: Verification using the entire data set
[0071] Accuracy verification was performed using data from the entire previously collected sample (151 individuals). Cross-validation was then performed to enhance the objectivity of the verification.
[0072] Verification Step 2: 7:3 Cross-Validation
[0073] A validation study was conducted to establish objective accuracy assessment and criteria. The entire data set was divided into a train set and a test set at a 7:3 ratio. To avoid data bias and underfitting, 70% of the data for each species was randomly sampled without replacement to establish a classification criterion, and the remaining 30% was used for accuracy verification. However, the LBS range, the primary classification criterion, was applied based on the reference range for the entire 151 individuals.
[0074] Verification step 3: 7:3 cross-validation after integration of sister species (Sispter sp.)
[0075] Among the nine species of the Calliphoridae collected, the sister species (Sispter sp.) Lucilia caesar (ca) and Lucilia illustris (il) were grouped into one species (ca+il), and the black-cheeked ...
[0076]
[0077] 3. Verification of the suitability of classification indicators
[0078] To assess the suitability of the classification indices used in the method for identifying black flies of the family Phormia according to the present invention, statistical validation was performed. For validation, 454 mature third-instar larvae randomly collected from black flies (Phormia regina) obtained through rearing experiments were analyzed. The analysis was performed using the statistical program R (ver. 4.2.3), and the correlation between individual classification indices and body length was analyzed to determine the association between individual classification indices and larval growth (increase in body length).
[0079]
[0080] Example 1. Analysis of basic data for classification indicators
[0081] The classification indices were measured in 9 species of flies in the family Black Fly, and organized by species for ABS_1-3 (°, angle), DBS (ratio), and LBS (mm). 12 to 25 samples were used for each species, and boxplots graphs corresponding to each classification indices were created based on the species-specific statistical data for the classification indices. As shown in Fig. 2, it was confirmed that there was a certain range for each classification indices for the 5 species of Black Fly (ABS_1, ABS_2, ABS_3, LBS, and DBS) depending on the species of Black Fly. In the case of ABS_1, ABS_2, and ABS_3, species belonging to the same subfamily showed similar patterns. Specifically, the large black fly (Calliphora lata) (la), the red-cheeked black fly (Calliphora vicina) (vi), and the two-tailed black fly (Triceratopyga calliphoroides) (cl) belonging to the Calliphorinae subfamily showed similar patterns, the green-headed fly (Lucilia illustris) (il), the golden fly (Lucilia caesar) (ca), and the copper-headed fly (Lucilia sericata) (se) belonging to the Luciliinae subfamily showed similar patterns, and the black-cheeked black fly (Chrysomya megacephala) (mg), the large black-cheeked black fly (Chrysomya pinguis) (pi), and the black-headed black fly (Phormia regina) (re) belonging to the Chrysomyinae subfamily showed similar patterns. The LBS index was able to separate all target species into four groups according to their range, but because the standard deviation range was smaller than that of other indices, it was confirmed to be the most suitable classification index for classifying nine species of the Calliphoridae family in the first classification stage.
[0082]
[0083] Example 2. Application of unknown samples
[0084] The method for identifying black flies of the family Calliphoridae according to the present invention was applied to an unknown sample to identify the species of black flies of the family Calliphoridae. The unknown sample was a separately collected sample (an individual unrelated to the previous experimental example, the individual of Figure 3), and the measurement values for five classification indices (ABS_1, ABS_2, ABS_3, LBS, and DBS) were obtained from the unknown sample as shown in Table 9 below.
[0085] Unknown sampleABS_171.70°ABS_297.56°ABS_3129.55°LBS (R)0.1853 μmLBS (L)0.1912 μmLBS (average)0.1883 μmDistance0.4331 μmLBS0.1883 μmDBS2.3001
[0086]
[0087] Using the measurement values of the classification index of the unknown sample measured above, the fly species was identified according to the species identification method of the Calliphoridae fly according to the present invention. In the first classification step, the LBS measurement value of the unknown sample was confirmed to be 0.18295 μm, and the red-cheeked black fly (Calliphora vicina) (vi) (over 0.1475 mm or less or 0.19 mm), the green-fly (Lucilia illustris) (il) (over 0.17 mm or less or 0.20 mm), the golden fly (Lucilia caesar) (ca) (over 0.18 mm or less or 0.22 mm), and the copper-fly (Lucilia sericata) (se) (over 0.17 mm or less or 0.21 mm), which correspond to the LBS reference range to which the LBS measurement value of the unknown sample belongs, were selected as the first classification reference species. In the second classification step, the classification index error rates of unknown samples were calculated and added up with respect to the reference values of the classification indexes for the above four species, as shown in Table 10 below. The sum of the error rates was 0.9007 for the red-cheeked black fly (Calliphora vicina) (vi), 0.8917 for the green-green golden fly (Lucilia illustris) (il), 1.1571 for the golden fly (Lucilia caesar) (ca), and 0.3014 for the copper golden fly (Lucilia sericata) (se). Finally, the unknown sample was identified as the copper golden fly (Lucilia sericata) (se) with the lowest sum of the error rates.
[0088] Classification laviclilcasemgpireLBS Range0.22<0.1475<0.12<0.17<0.18<0.17<0.28<0.27<0.23<≤0.26≤0.19≤0.1475≤0.20≤0.22≤0.21≤0.37≤0.36≤0.27ABS_1Error rate0.32260.08630.00870.0562ABS_2Error rate0.23 930.10940.07870.0195ABS_3Error rate0.19980.11520.11320.0424DBSError rate0.02270.57820.87680.1646LBSError rate0.11640.00280.07970.0187Total error rate0.90070.89171.15710.3014
[0089]
[0090] Example 3. Results of verification of the accuracy of the classification method
[0091] Example 3-1. Verification using full data
[0092] The above verification was conducted using the entire previously collected sample (151 individuals) as is and in the same manner as Example 2. In the above experimental example, the accuracy of the present invention was measured by measuring the classification index for 151 individuals whose species were previously confirmed. As shown in Table 11 below, the species identification result for all 151 individuals showed an accuracy of 92.7% (140 / 151). Among the 9 species, some of the sister species, Lucilia caesar (ca) and Lucilia illustris (il), were identified inconsistently, showing an accuracy of 86.8% (33 / 38). Some of the sister species, Chrysomya megacephala (mg) and Chrysomya pinguis (pi), were identified inconsistently, showing an accuracy of 82.8% (29 / 35). The remaining five species showed 100% accuracy.
[0093] Identification Results Actual Species Laviclibcasemgpirela21--------vi-18-------cl--14------il---222----ca---311----se-----13---mg------143-pi------315-re--------12
[0094]
[0095] Example 3-2. 7:3 Cross-Validation
[0096] In the above experiment, 70% of the data randomly extracted without replacement from each species from 151 individuals were used to reset the reference values of the classification indicators, and the accuracy was measured using the remaining 30% of the data in the same manner as in Example 2. The entire process was repeated 1,000 times, and the results showed an average accuracy of 91.0%.
[0097]
[0098] Example 3-3. 7:3 Cross-Validation after Integration of Sister Species
[0099] Sister species (Sispter sp.) Lucilia caesar (ca) and Lucilia illustris (il) were grouped into one species (ca+il), and black-cheeked ...
[0100]
[0101] Example 4. Correlation Analysis of Classification Indicators
[0102] The body length (Length), ABS_2, DBS, and LBS of 454 black goldflies (Phormia regina) were investigated using the statistical program R (ver. 4.2.3), as shown in Table 12 below.
[0103] P. reginameansdminmaxmedianLength14.461.1612.2117.2114.43ABS_282.788.9056.40103.0082.72DBS1.670.181.112.191.67LBS (pixel)120.778.66100.10158.28120.55
[0104]
[0105] In order to analyze the correlation between each indicator, the correlation coefficient of each indicator value was measured and created as a heatmap, and the distribution pattern of ABS_2, DBS, and LBS with respect to body length was created as a scatter plot. As shown in Fig. 4, the correlation coefficient of ABS_2 with respect to body length was 0.0936, that of DBS was 0.2965, and that of LBS was -0.0342. In particular, LBS showed a correlation coefficient close to 0, confirming that it is an indicator unrelated to body length. Although ABS and DBS showed a low positive correlation with body length, they showed a low correlation of less than 0.5, confirming their validity as indicators for species identification. The distribution pattern of ABS_2, DBS, and LBS with respect to body length in the scatter plot was found to be scattered without a specific pattern. The above results demonstrate that the morphological classification index of the latter sentence has little correlation with the degree of fly growth, and that the fly classification method according to the present invention is a classification method capable of identifying species without being affected by the growth process.
[0106] While specific aspects of the present invention have been described in detail above, it should be apparent to those skilled in the art that these specific descriptions merely represent preferred embodiments and are not intended to limit the scope of the present invention. In other words, the substantial scope of the present invention is defined by the appended claims and their equivalents.
[0107] Numerical ranges are inclusive of the values defined in the ranges above. Any maximum numerical limitation given throughout this specification includes any lower numerical limitation, as if that lower numerical limitation were explicitly stated. Any minimum numerical limitation given throughout this specification includes any higher numerical limitation, as if that higher numerical limitation were explicitly stated. Any numerical limitation given throughout this specification will include any better numerical range within the broader numerical range, as if that narrower numerical limitation were explicitly stated.
Claims
In the fly identification system using the morphological characteristics of the 1.3 instar larvae, An image collection unit that collects images of the posterior spiracle by photographing the posterior spiracle of third instar larval from a group of flies; A data collection unit that extracts only the posterior pronotum images of a standard fly population whose scientific name has been identified among the above fly populations, and measures morphological characteristics from the posterior pronotum images of the extracted standard fly populations to collect morphological data; A data processing unit for calculating reference values of the morphological data, including an average value, a standard deviation value, a minimum value, a maximum value, and a median value statistically calculated from the collected morphological data; A classification criterion model construction unit that constructs a first classification criterion model and a second classification criterion model from the above-described calculated criterion values and classifies them according to the scientific name of the standard fly; A measurement value collection unit that extracts only the posterior pronotum images of unknown fly populations whose scientific names are unknown among the above fly populations, and collects measurement values by measuring morphological characteristics from the posterior pronotum images of the extracted unknown fly populations; and A fly identification system including an analysis unit that compares the reference values provided from the first classification reference model and the second classification reference model with the measured values to identify the fly.
2. In paragraph 1, The above morphological features are, ABS_1, the angle (°) formed by the first pair of slits (US, Upper slits) located at the top of the three pairs of slits symmetrically placed on both sides of the back door; ABS_2, the angle (°) formed by the second pair of slits (MS, Middle Slits) located in the middle of the three pairs of slits symmetrically placed on both sides of the back door; ABS_3, the angle (°) formed by the third pair of slits (LS, Lower Slits) located at the bottom of the three pairs of slits symmetrically placed on both sides of the back door; LBS (Length BiSlit), which is the average length (mm) of the second slit; and A Paris identification system characterized by including DBS (Distance Between Slit), which is a value obtained by dividing the shortest distance (mm) between the lower points of the second slits symmetrical on both sides of the back door by the LBS.
3. In paragraph 1, A Paris identification system characterized in that the first classification criterion model is composed of ranges expressed as minimum and maximum values of the morphological data.
4. In paragraph 1, A Paris identification system characterized in that the second classification criterion model is composed of median values of the morphological data. A method for identifying flies using a fly identification system that utilizes the morphological characteristics of 5.3 instar larvae. A step of collecting images of the posterior larvae obtained by photographing the posterior larvae of unknown, unidentified third-instar flies; A step of identifying three pairs of slits symmetrical on both sides from the collected image, and recognizing the slit pairs located at the topmost from the center of symmetry as the first slit pair, the second slit pair, and the third slit pair; A step of measuring the angle (°) formed by the three pairs of slits, and measuring the angle (°) value formed by the first slit pair, ABS_1, the angle (°) value formed by the second slit pair, and ABS_3, the angle (°) value formed by the third slit pair; A step of measuring the length of each of the second slit pairs and measuring the average length (mm) value, LBS (Length BiSlit); A step of measuring DBS (Distance Between Slit), which is the shortest distance (mm) between the lower points of the second slits divided by the LBS; and A method for identifying a fly with a fly identification system, comprising: a step of identifying a fly by comparing the measurement values of ABS_1, ABS_2, ABS_3, LBS, and DBS with the first classification reference models and the second classification reference models constructed according to the scientific name of the fly in the fly identification system.
6. In paragraph 5, The steps to identify the above fly are: A step of comparing the LBS measurement values with the first classification reference models and selecting the scientific name of the fly corresponding to the first classification reference model including the LBS measurement values as the classification reference species; A step of calculating the absolute error rate by comparing the reference values of the second classification criterion models of the above-mentioned selected classification criterion species with the measured values of ABS_1, ABS_2, ABS_3, LBS, and DBS; and A method for identifying a fly using a fly identification system, characterized in that it comprises a step of identifying the unknown fly as a classification criterion species in which the sum of the absolute values of the calculated error rates is the lowest.
7. In paragraph 6, In the step of selecting the above classification criterion species, if only one first classification criterion model is selected, A method for identifying a fly using a fly identification system characterized in that the unknown fly is identified by the scientific name of a standard fly corresponding to the first classification criterion model selected above.
8. In paragraph 6, A method for identifying flies using a fly identification system, characterized in that the above absolute error rate is a value calculated by substituting the reference value of the second classification reference model and the measured value into [Mathematical Formula 1] below. [Mathematical Formula 1] In a method for species identification of Calliphoridae flies using morphological characteristics of 9.3 instar larvae, (1) A step of preparing an image by photographing the posterior spiracle of the third instar larval of the fly to be identified; (2) A step of confirming three pairs of slits symmetrical on both sides of the back text in the image above, and recognizing the slit pairs located from the center of symmetry to the topmost pair as the first slit pair, the second slit pair, and the third slit pair; (3) A step of measuring the length of each of the second slit pairs and measuring the average length (mm) value, LBS (Length BiSlit); and (4) A method for identifying a species of a Calliphoridae fly, comprising the step of selecting a classification reference species corresponding to an LBS reference range including the measured value of the LBS by comparing the first classification reference model of the Calliphoridae fly shown in Table 1 below. [Table 1] 10. In paragraph 9, if two or more species based on the classification criteria are selected, (4) A step of measuring ABS_1, which is an angle (°) value formed by the first slit pair, ABS_2, which is an angle (°) value formed by the second slit pair, ABS_3, which is an angle (°) value formed by the third slit pair, and DBS (Distance Between Slit), which is a value obtained by dividing the shortest distance (mm) between the lower points of the second slits by the LBS; (5) A step of calculating the absolute error rate by comparing the reference values provided in the second classification reference model of the black fly (Calliphoridae) shown in Table 2 below with the measured values of ABS_1, ABS_2, ABS_3, LBS, and DBS; and (6) A method for identifying a species of a fly of the Calliphoridae family, characterized in that it further includes a step of identifying the fly as a classification criterion species in which the sum of the calculated absolute values of the error rates is the lowest. [Table 2] 11. In paragraph 10, A method for identifying a species of a Calliphoridae fly, characterized in that the above absolute error rate is a value calculated by substituting the reference value of Table 2 and the measured values of ABS_1, ABS_2, ABS_3, LBS, and DBS into [Mathematical Formula 1] below. [Mathematical Formula 1]
Citation Information
Patent Citations
Insect image processing apparatus, image processing method, image processing program, and computer-readable storage medium
JP2012161269A
Captured insect identification method and identification system
JP2014142833A
System for recognizing vermin and method thereof
KR1020090049933A
Estimation of Minimum Postmortem Interval Using Developmental Gene Expression of Necrophagous Fly Species
KR102262568B1
System and method for providing a voice data management platform with NFT technology applied
KR102486563B1
