Diatom morphological classification method for forensic medicine drowning place inference

Abstract

The invention relates to the technical field of morphological classification, and discloses a diatom morphological classification method for forensic medicine drowning place inference, and the method comprises the steps: obtaining diatom vector data and organization vector data; respectively extracting three principal component variables from the diatom vector data and the tissue vector data to obtain a first coordinate point and a second coordinate point; calculating the distance between the first coordinate point and the second coordinate point corresponding to each sampling point; and taking the sampling point with the shortest distance from the first coordinate point as a potential drowning place. According to the application, the principal component analysis is carried out on the tissue vector data and the diatom vector data to obtain the first coordinate point and the second coordinate points formed by the three principal component variables, and the sampling point of the second coordinate point with the shortest distance to the first coordinate point is used as the potential drowning place according to the distance between the first coordinate point and each second coordinate point. By means of the mode, drowning place inference which is rapid, accurate and easy to operate can be achieved.

Description

Technical Field

[0001] This application relates to the technical field of morphological classification, and particularly relates to a diatom morphological classification method for inferring the drowning location in forensic medicine. Background Art

[0002] In forensic medicine practice, inferring the drowning location is a crucial link in clarifying the circumstances of water-related death cases, especially when the location where the sunken body is found is far from the initial drowning location. The presumed drowning location is of great value in verifying or refuting eyewitness testimony, accurately estimating the time interval after death, and revealing environmental factors that may have led to the accident. In addition, accurately inferring the drowning location also makes a significant contribution to public safety work, which can help identify high-risk areas for water accidents. This valuable information supports targeted prevention strategies, such as optimizing the setting of safety signs, strategically arranging lifeguard posts, and conducting comprehensive public education and awareness campaigns. By leveraging these insights, relevant departments can strengthen water safety measures and potentially reduce the number of drowning-related deaths.

[0003] Diatoms are single-celled algae with unique siliceous cell walls and are widely distributed in aquatic environments. During the drowning process, these microorganisms may be inhaled into the alveolar cavity, then enter the blood circulation through the alveolar capillaries, and are distributed in organs such as the liver, kidneys, and bone marrow. In some countries, the diagnosis of drowning heavily relies on the detection of diatoms in these organs (especially the lungs, liver, and kidneys), although the reliability of diatom detection has always been controversial. In addition, forensic experts can infer the drowning location by comparing the diatom taxa found in the organs with those in water samples. Currently, some scholars have used molecular biology techniques to diagnose drowning and infer the drowning location by identifying diatom species, specifically using DNA barcoding techniques such as 16S rDNA, 18S rDNA, and rbcL. However, these methods have been questioned due to the generation of non-specific products during the PCR amplification process. In addition, due to the high cost of instruments and labor, the wide application of these methods in forensic medicine is hindered.

[0004] In contrast, the morphological definition of diatom species provides a more economical alternative, requiring less specialized equipment than molecular techniques and being more cost-effective, which facilitates its popularization at the forensic grass-roots level. The morphological identification and classification of diatoms involve systematically examining diatom frustules under a high magnification lens (>1000×), usually using an oil immersion lens or an electron microscope. The traditional process generally follows the following steps: (1) identifying the presence of diatom frustules; (2) distinguishing key physical characteristics, such as diameter and the direction of the suture (centered or eccentric); (3) comparing characteristics, including the number of striae rows, frustule valve length, stria diameter, and width; (4) verifying symmetry after identifying the diatoms to confirm whether it is bilateral symmetry or radial symmetry. Other characteristics, such as shape and size, should also be analyzed to further refine the classification. This method requires a considerable amount of expertise to distinguish subtle morphological differences between species and usually requires referring to taxonomic atlases and other materials to ensure accuracy. This method is extremely complex, time-consuming, and requires a great deal of aquatic knowledge, especially in time-sensitive applications such as forensic investigations. Currently, there is an urgent need to develop faster and easier-to-operate diatom classification techniques for use in forensic drowning site inference. Summary of the Invention

[0005] The purpose of this application is to provide a diatom morphological classification method for forensic drowning site inference to achieve fast, accurate, and easy-to-operate drowning site inference.

[0006] To achieve the above purpose, the following technical solutions are adopted:

[0007] This application provides a diatom morphological classification method for forensic drowning site inference, and the method includes:

[0008] Obtaining diatom vector data and tissue vector data;

[0009] Extracting three principal component variables from the diatom vector data and the tissue vector data respectively, and representing the three extracted principal component variables as coordinate points in three-dimensional space to obtain a first coordinate point and a second coordinate point;

[0010] Calculating the distance between the first coordinate point and the second coordinate points corresponding to each sampling point;

[0011] Taking the sampling point with the shortest distance from the first coordinate point as the potential drowning site.

[0012] Furthermore, the diatom vector data and the tissue vector data are obtained in the following manner:

[0013] Collecting water sample specimens from multiple sampling points in the target area and making them into water sample smears;

[0014] Counting the number of diatoms in the water sample smears of each sampling point;

[0015] Classify the diatoms in the water sample smear using a diatom classification system and calculate the percentage of each type of diatom in the total number of diatoms;

[0016] Arrange the percentage values of each type of diatom in a set order to obtain the diatom vector data for each sampling location;

[0017] Extract a tissue sample from the human body to be tested, make it into a tissue smear, and classify it using a diatom classification system to obtain tissue vector data.

[0018] Furthermore, the number of diatoms in the water sample smear at each sampling point is not less than 500.

[0019] Furthermore, the method of classifying the diatoms in the water sample smear or the tissue sample using a diatom classification system includes:

[0020] Extract the diatom shell characteristics, contour characteristics, and texture characteristics from the water sample smear or the tissue sample;

[0021] Determine the symmetry classification of the diatoms according to the diatom shell characteristics;

[0022] Determine the contour classification of the diatoms according to the contour characteristics;

[0023] Determine the texture classification of the diatoms according to the texture characteristics.

[0024] Furthermore, the method of classifying the diatoms in the water sample smear or the tissue sample using a diatom classification system also includes:

[0025] Represent the symmetry classification, contour classification, and texture classification of the diatoms as a three-dimensional variable, where the three-dimensional variable includes three variable values, and the three variable values respectively represent the numbers of the symmetry classification, contour classification, and texture classification of the diatoms.

[0026] Furthermore, determining the symmetry classification of the diatoms according to the diatom shell characteristics includes:

[0027] If the diatom shell characteristics are that the diatom shell rotates around a central point at any angle while maintaining its shape and appearance unchanged, then determine the symmetry classification as central symmetry;

[0028] If the diatom shell characteristics are that the diatom shell shows symmetry on both the apical axis and the transverse apical axis, and remains unchanged when rotated at a specific angle less than 360 degrees, then determine the symmetry classification as bilateral symmetry;

[0029] If the diatom shell characteristics are that the diatom shell has an inverted part aligned along the apical axis or the transverse apical axis, and remains unchanged when rotated at a certain angle less than 360 degrees, then determine the symmetry classification as antisymmetry;

[0030] If the diatom shell feature shows symmetry on the apical axis or the transverse apical axis, such that it remains unchanged after being rotated by an angle less than 360 degrees, then determine that the symmetry classification is single-axis symmetry;

[0031] If the diatom shell feature is asymmetric on the apical axis and the transverse apical axis, and it does not appear the same after being rotated by any angle less than 360 degrees, then determine that the symmetry classification is asymmetry.

[0032] Furthermore, determine the contour classification of the diatom according to the contour feature, including:

[0033] If the contour classification is a circular two-dimensional figure where all points are equidistant from the center point, then determine that the contour classification is circular;

[0034] If the contour classification is a two-dimensional elliptical curve where the sum of the distances from any point on the curve to two fixed points remains constant, then determine that the contour classification is elliptical;

[0035] If the contour classification is an elongated symmetric shape with tapered ends and a bulging side, then determine that the contour classification is spindle-shaped;

[0036] If the contour classification is a slender shape with tapering ends and relatively straight sides, then determine that the contour classification is needle-shaped;

[0037] If the contour classification is a slender, straight rectangle with flat ends, then determine that the contour classification is rod-shaped;

[0038] If the contour classification is a smooth, flowing line or shape without sharp angles, which can be open or closed and may gradually change direction, then determine that the contour classification is curvilinear;

[0039] If the contour classification is a two-dimensional shape with four sides and four right angles, then determine that the contour classification is square;

[0040] If the contour classification is a slender, extended shape with sharp ends, then determine that the contour classification is willow-leaf-shaped;

[0041] If the contour classification is a boat-shaped shape characterized by a slender, curved structure, wider in the middle and tapering at both ends, then determine that the contour classification is crescent-shaped;

[0042] If the contour classification is a slender shape with two rounded and bulging ends connected by a narrower middle part, then determine that the contour classification is peanut-shaped;

[0043] If the contour is classified as a symmetric shape with a circular or elliptical center and twisted ends, then determine that the contour is classified as a candy shape;

[0044] If the contour is classified as a smooth, teardrop-shaped form with a rounded bottom and a gradually tapering top, then determine that the contour is classified as a water droplet shape;

[0045] If the contour is classified as a shape consisting of two parallel, slender, equal-length, and uniformly wide rods, then determine that the contour is classified as a double rod shape;

[0046] If the contour is classified as having a sharp triangular outline, with a broad bottom that gradually narrows to a sharp tip and a distinct protrusion at the bottom, then determine that the contour is classified as a dart shape.

[0047] Furthermore, determining the texture classification of the diatom according to the texture feature includes:

[0048] If the texture feature is without any texture, then determine that the texture classification of the diatom is an empty texture;

[0049] If the texture feature is a pattern with equal, short, fence-like structures along the edge, and the fence-like structures are located on the inner or outer side, then determine that the texture classification of the diatom is a serrated texture;

[0050] If the texture feature is a pattern in which straight, uniform lines run through the entire body at equal intervals, forming a striped or linear appearance, then determine that the texture classification of the diatom is a striped texture;

[0051] If the texture feature is a pattern in which straight, uniform lines run through the entire body at equal intervals, forming a striped or linear appearance, then determine that the texture classification of the diatom is a vertical striped texture;

[0052] If the texture feature is a pattern in which a prominent single vertical line runs through the entire body from top to bottom and is located at the central position, then determine that the texture classification of the diatom is a cross-shaped texture;

[0053] If the texture feature is a pattern defined by the intersection of uniformly distributed horizontal lines and a single vertical line running from top to bottom, then determine that the texture classification of the diatom is a fence-like texture;

[0054] If the texture feature is a pattern consisting of two concentric circles, and the edges of each circle may be decorated with a radial pattern, then determine that the texture classification of the diatom is a concentric circle texture.

[0055] Furthermore, the method of respectively extracting three principal component variables from the diatom vector data and the tissue vector data includes respectively extracting the first three principal component variables from the diatom vector data and the tissue vector data based on principal component analysis.

[0056] Further, the method for calculating the distance between the first coordinate point and the second coordinate points corresponding to each sampling point includes cosine distance or Euclidean distance.

[0057] The beneficial effects of this application are as follows:

[0058] This application uses a diatom classification system to classify diatoms in tissue smears and water smears at each sampling point, obtaining tissue vector data and diatom vector data, and performing principal component analysis on the tissue vector data and diatom vector data to obtain a first coordinate point and a second coordinate point formed by three principal component variables. According to the distance between the first coordinate point and each second coordinate point, the sampling point where the second coordinate point with the shortest distance to the first coordinate point is located is used as the potential drowning location. Through the above method, this application can achieve fast, accurate, and easy-to-operate drowning location inference. Description of the Drawings

[0059] Figure 1 It is a flowchart of a diatom morphological classification method for forensic drowning location inference provided by an embodiment of this application;

[0060] Figure 2 It is a flowchart of classifying diatoms in a water smear or tissue sample using a diatom classification system provided by an embodiment of this application;

[0061] Figure 3 It is a schematic diagram of symmetry classification provided by an embodiment of this application;

[0062] Figure 4 It is a schematic diagram of contour classification provided by an embodiment of this application;

[0063] Figure 5 It is a schematic diagram of texture classification provided by an embodiment of this application;

[0064] Figure 6 It is an effect diagram of a diatom morphological classification method for forensic drowning location inference provided by an embodiment of this application; wherein, A, PCA scatter plot of in vitro experiment; B, PCA scatter plot of in vivo experiment; C, confusion matrix diagram of drowning location inference in vitro experiment; D, confusion matrix diagram of drowning location inference in vitro experiment. Detailed Embodiments

[0065] The following uses specific specific examples to illustrate the implementation manners of the present application. Those skilled in the art can easily understand other advantages and effects of the present application from the content disclosed in this specification. The present application can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0066] The following further describes in detail the specific implementation manners of the present application in conjunction with the accompanying drawings and embodiments.

[0067] Please refer to Figure 1 , which is a flowchart of a diatom morphological classification method for forensic drowning location inference provided by an embodiment of the present application. An embodiment of the present application provides a diatom morphological classification method for forensic drowning location inference. This method can be implemented by the following steps S100 to S103.

[0068] S100: Obtain diatom vector data and tissue vector data.

[0069] In some embodiments, step S100 includes the following steps S101 to S105.

[0070] S101: Collect water sample specimens at multiple sampling points in the target area and make them into water sample smears;

[0071] S102: Count the number of diatoms in the water sample smear of each sampling point;

[0072] S103: Classify the diatoms in the water sample smear using a diatom classification system and calculate the percentage of each type of diatom in the total number of diatoms;

[0073] S104: Arrange the percentage values of each type of diatom in a set order to obtain the diatom vector data of each sampling location;

[0074] S105: Extract a tissue sample to be tested and make it into a tissue smear, and classify it using a diatom classification system to obtain tissue vector data.

[0075] S101: Respectively extract three principal component variables from the diatom vector data and the tissue vector data, and represent the three principal component variables extracted as coordinate points in three-dimensional space to obtain a first coordinate point and a second coordinate point.

[0076] S102: Calculate the distance between the first coordinate point and the second coordinate points corresponding to each sampling point.

[0077] S103: Use the sampling point with the shortest distance from the first coordinate point as the potential drowning location.

[0078] In some embodiments, such as Figure 2 shown, the method for classifying diatoms in the smear of the water sample or the tissue sample by using the diatom classification system includes the following steps:

[0079] S201: Extract the diatom shell features, contour features, and texture features from the smear of the water sample or the tissue sample;

[0080] S202: Determine the symmetry classification of the diatoms according to the diatom shell features;

[0081] S203: Determine the contour classification of the diatoms according to the contour features;

[0082] S204: Determine the texture classification of the diatoms according to the texture features.

[0083] In this embodiment, the diatom classification system can realize the classification of diatoms in human tissues and water sample smears under a conventional microscope within 400 times magnification. It identifies diatoms through the diatom shell features, contour features, and texture features extracted from the smear. Each diatom can be defined by a three-dimensional vector. For example, the three-dimensional vector can be (a, b, c), where a, b, and c are respectively a vector value, representing the symmetry classification number, contour classification number, and texture classification number of the diatom.

[0084] Specifically, as Figure 3 shown, Figure 3 in which, Radial represents central symmetry, Double-axis represents bilateral symmetry, Single-axis represents single-axis symmetry, Asymmetry represents asymmetry, Symmetry represents symmetry, and Example represents an example. Regarding the symmetry classification, it is divided into five types, and the corresponding symmetry classification number a is any natural number from 1 to 5. Specifically as follows:

[0085] 1) Central symmetry: The diatom shell can rotate around a central point at any angle while maintaining its shape and appearance unchanged;

[0086] 2) Bilateral symmetry: The diatom shell shows symmetry on both the apical axis and the transverse apical axis, making it remain unchanged when rotated at a specific angle less than 360 degrees;

[0087] 3) Reverse symmetry: The diatom shell has an inverted part aligned along the apical axis or the transverse apical axis, making it remain unchanged when rotated at a specific angle less than 360 degrees;

[0088] 4) Single-axis symmetry: The diatom shell shows symmetry on the apical axis or the transverse apical axis, making it appear to remain unchanged after being rotated at a specific angle less than 360 degrees;

[0089] 5) Asymmetry: The diatom frustules are asymmetric along the apical and transapical axes, which makes them not appear the same after rotation at any angle less than 360 degrees.

[0090] As shown in Figure 4 the figure, Figure 4 where Circle represents circular, Ellipse represents oval, Spindle represents fusiform, Needle represents acicular, Rod represents rod-shaped, Curve represents curved, Square represents square, Outline represents contour, Example represents example, Willow leaf represents lanceolate, Navicular represents crescent, Peanut represents peanut-shaped, Candy represents candy-shaped, Drop represents drop-shaped, Double-bar represents double-bar-shaped, Dart represents dart-shaped. Regarding the contour classification, it is mainly divided into 14 types, and the corresponding contour classification number b is any natural number from 1 to 14, specifically as follows:

[0091] 1) Circular: A perfect two-dimensional circular figure where all points are equidistant from the center point;

[0092] 2) Oval: A two-dimensional oval curve where the sum of the distances from any point on the curve to two fixed points remains constant;

[0093] 3) Fusiform: An elongated symmetric shape that tapers at both ends and bulges on the sides;

[0094] 4) Acicular: A slender shape that tapers at both ends and has relatively straight sides;

[0095] 5) Rod-shaped: A slender and straight rectangle with flat ends;

[0096] 6) Curved: A smooth and flowing line or shape without sharp angles, which can be open or closed and may gradually change direction;

[0097] 7) Square: A two-dimensional shape with four sides and four right angles;

[0098] 8) Lanceolate: An elongated and extended shape similar to a willow leaf, with sharp ends and often gently curved like the letter 'S';

[0099] 9) Crescent: A boat-shaped shape characterized by a slender and curved structure, wider in the middle and tapering at both ends;

[0100] 10) Peanut-shaped: A slender shape with two rounded and bulging ends connected by a narrower middle, similar to the shape of the number '8';

[0101] 11) Candy-like: A small, symmetrical shape with a circular or oval center and twisted ends at both sides, similar to a wrapped candy;

[0102] 12) Droplet-like: A smooth, teardrop-shaped form with a rounded bottom and a gradually tapering top;

[0103] 13) Double-rod shape: A shape composed of two parallel, slender, equal-length rods with a uniform width;

[0104] 14) Dart shape: A shape similar to a Chinese dart, with a sharp triangular outer shape, a wide bottom that gradually narrows to a sharp tip, and a distinct protrusion at the bottom.

[0105] As Figure 5 shown in the figure, in the figure, Empty represents an empty texture, Serrated represents a serrated texture, Strip represents a stripe texture, VD represents a vertical line texture, CD represents a cross-shaped texture, Fence-like represents a fence-like texture, CC represents a concentric circle texture, and Pattern represents a texture. Regarding texture classification, it is mainly divided into 7 types, and the corresponding texture classification number c is any natural number from 1 to 7. Specifically as follows:

[0106] 1) Empty texture: Without any texture;

[0107] 2) Serrated texture: A pattern with equal, short, fence-like structures along the edge, which can be located inside or outside;

[0108] 3) Stripe texture: A pattern in which straight, uniform lines run through the entire body at equal intervals, forming a stripe or linear appearance;

[0109] 4) Vertical line texture: A pattern in which straight, uniform lines run through the entire body at equal intervals, forming a stripe or linear appearance;

[0110] 5) Cross-shaped texture: A pattern with a prominent single vertical line running through the entire body from top to bottom and located in the central position;

[0111] 6) Fence-like texture: A pattern defined by evenly distributed horizontal lines intersecting with a single vertical line running from top to bottom;

[0112] 7) Concentric circle texture: A pattern composed of two concentric circles, and the edges of each circle may be decorated with a radial pattern.

[0113] In an exemplary embodiment, the diatom water sample database of the target area can be constructed through steps S101 to S104 in the process as Figure 1 shown, and then the drowning location inference can be achieved through the following steps:

[0114] (1) Extract the human tissue sample to be tested to make a smear and use the diatom classification system to form tissue vector data;

[0115] (2) Use principal component analysis to analyze the database and the human tissue vector data respectively, and extract the first three principal component variables. These variable values can be represented as coordinate points of (x, y, z) in three-dimensional space;

[0116] (3) Calculate the distances between the coordinate points formed by the three principal component values of the human tissue sample to be tested and the coordinate points formed by the three principal component values of each sampling water sample point in the database;

[0117] (4) The water sample sampling point with the shortest distance to the human tissue sample to be tested is the potential drowning location.

[0118] The final result obtained is as Figure 6 shown.

[0119] Figure 6 In A - B in, the 3D scatter plots generated after principal component analysis (PCA) of the diatom abundance data of four sampling points (CF, ZS, CY, and TS) are shown, revealing the diatom community characteristics of different water areas. In these figures, the large dots represent the centroids of the reference water sample samples, while the small dots correspond to the samples in in vitro experiments and in vivo experiments. The PCA results show that the diatom communities at different sampling points form distinct and independent clusters, indicating that the water source has a significant impact on the diatom community, making it an effective marker for forensic water identification. For example, the samples at the CY sampling point form a tight and independent cluster, indicating that its unique diatom composition can distinguish CY from other sampling points (CF, ZS, and TS) (including the samples in in vitro experiments and in vivo experiments).

[0120] In addition, the centroid (large dot) of each cluster provides a central reference point for distance - based classification analysis. Specifically, these centroids can be used to calculate the Mahalanobis distance, so as to classify unknown samples into the most likely drowning locations. By comparing the distances from each unknown sample in in vitro experiments and in vivo experiments to the nearest centroid, it can be associated with the corresponding water source with high confidence. Based on the calculation of the Mahalanobis distance, both in vitro experiments and in vivo experiments show high classification accuracy in drowning location identification. Among them, the classification accuracy of the in vitro experiment (n = 44) reaches 0.98, indicating that almost all samples can be accurately identified and attributed to the water area; in contrast, the classification accuracy of the in vivo experiment (n = 40) is slightly lower, at 0.95. In addition, Figure 6C-D in the figure shows the confusion matrix of the classification results of in vitro and in vivo experiments, showing high classification accuracy for three sampling points (CY, ZS, and TS), and all samples at these sampling points were correctly classified. However, there were slight misclassifications at the CF sampling point in both confusion matrices: in the in vitro experiment ( Figure 6 C), 1 CF sample was misclassified as CY; while in the in vivo experiment ( Figure 6 D), 2 CF samples were misclassified as TS.

[0121] The above embodiments are only used to illustrate the present application and are not intended to limit the present application. Those of ordinary skill in the relevant technical field can also make various changes and modifications without departing from the spirit and scope of the present application. Therefore, all equivalent technical solutions also belong to the scope of the present application, and the patent protection scope of the present application shall be defined by the claims.

Claims

1. A diatom morphological classification method for forensic drowning location inference, characterized in that: The method comprises: Obtaining diatom vector data and tissue vector data; Extracting three principal component variables from the diatom vector data and the tissue vector data respectively, and representing the extracted three principal component variables as coordinate points in a three-dimensional space to obtain a first coordinate point and a second coordinate point; Calculating the distance between the first coordinate point and the second coordinate point corresponding to each sampling point; The sampling point with the shortest distance to the first coordinate point is taken as the potential drowning location.

2. The diatom morphological classification method for forensic drowning location inference according to claim 1, characterized in that: The diatom vector data and tissue vector data are obtained by the following methods: Collect water samples from multiple sampling points in the target area and make water smears; The number of diatoms in the water smears from each sampling point was counted; Using the diatom classification system, classify the diatoms in the water sample smear and calculate the percentage of each diatom in the total number of diatoms; Arrange the percentage values ​​of each diatom in the set order to obtain the diatom vector data of each sampling site; The human tissue sample to be tested is extracted and made into a tissue smear, which is classified using the diatom classification system to obtain tissue vector data.

3. The diatom morphological classification method for forensic drowning location inference according to claim 2, characterized in that: The number of diatoms in the water sample smear at each sampling point shall be no less than 500.

4. The diatom morphological classification method for forensic drowning location inference according to claim 2, characterized in that: Methods for classifying the diatoms in the water sample or the tissue sample using the diatom classification system include: extracting diatom frustule features, contour features, and texture features from the water sample smear or the tissue sample; Determining the symmetry classification of diatoms based on the diatom frustule characteristics; determining the contour classification of the diatom according to the contour features; The texture classification of diatoms is determined according to the texture features.

5. The diatom morphological classification method for forensic drowning location inference according to claim 4, characterized in that: The method of classifying the diatoms in the water sample or the tissue sample using the diatom classification system also includes: The symmetry classification, outline classification and texture classification of diatoms are represented by a three-dimensional variable, wherein the three-dimensional variable includes three variable values, and the three variable values ​​represent the numbers of the symmetry classification, outline classification and texture classification of diatoms respectively.

6. The diatom morphological classification method for forensic drowning location inference according to claim 4, characterized in that: The symmetry classification of diatoms is determined according to the diatom frustule characteristics, including: If the diatom frustule is characterized by the diatom frustule being rotated at any angle around a central point while maintaining its shape and appearance unchanged, the symmetry classification is determined to be central symmetry; If the diatom frustule is characterized in that the diatom frustule exhibits symmetry on both the apical axis and the transverse apical axis, so that it remains unchanged when rotated at a specific angle less than 360 degrees, then the symmetry is determined to be classified as biaxial symmetry; If the diatom frustule is characterized by having an inverted portion aligned along an apical axis or a transverse apical axis so that it remains unchanged when rotated at an angle less than 360 degrees, the symmetry classification is determined to be inverse symmetry; If the diatom frustule is characterized in that the diatom frustule exhibits symmetry on the apical axis or the transverse apical axis, so that it remains unchanged after being rotated at an angle less than 360 degrees, then the symmetry is determined to be classified as uniaxial symmetry; If the diatom frustule is characterized in that the diatom frustule is asymmetric about the apical axis and the transapical axis, which does not appear the same after being rotated at any angle less than 360 degrees, then the symmetry classification is determined to be asymmetric.

7. The diatom morphological classification method for forensic drowning location inference according to claim 4, characterized in that: Determining the contour classification of diatoms according to the contour features includes: If the contour is classified as a circular two-dimensional figure and the distances from all points to the center point are equal, then the contour is determined to be classified as a circle; If the contour is classified as a two-dimensional elliptical curve, and the sum of the distances from any point on the curve to two fixed points remains constant, then the contour is determined to be classified as an ellipse; If the outline is classified as an elongated symmetrical shape with tapering ends and bulging sides, then the outline is determined to be classified as a spindle shape; If the outline is classified as an elongated shape with gradually tapering ends and relatively straight sides, then the outline is determined to be classified as a needle shape; If the contour is classified as an elongated, straight rectangle with flat ends, then the contour is determined to be classified as a rod; If the contour is classified as a smooth, flowing line or shape, has no sharp angles, may be open or closed, and may gradually change direction, then the contour is determined to be classified as a curvilinear shape; If the contour is classified as a two-dimensional shape with four sides and four right angles, then determining that the contour is classified as a square; If the outline is classified as a slender and extended shape with sharp ends, then the outline is determined to be classified as a willow leaf shape; If the outline is classified as a boat-shaped shape, characterized by a slender, curved structure, wider in the middle and tapering at both ends, then the outline is determined to be classified as a crescent shape; If the profile is classified as an elongated shape with two rounded and expanded ends connected by a narrower middle portion, then the profile is determined to be classified as a peanut shape; If the outline is classified as a symmetrical shape, with a circular or oval center and twisted ends, then the outline is determined to be classified as a candy shape; If the contour is classified as a smooth, teardrop-shaped shape with a rounded bottom and a gradually tapering top, then the contour is determined to be classified as a water drop shape; If the profile is classified as a shape consisting of two parallel, elongated rods of equal length and uniform width, then the profile is determined to be classified as a double rod shape; If the outline is classified as having a sharp triangular shape, with a wide base gradually narrowing to a sharp top, and a clear protrusion at the bottom, then it is determined that the outline is classified as a dart shape.

8. The diatom morphological classification method for forensic drowning location inference according to claim 4, characterized in that: Determining the texture classification of diatoms according to the texture features includes: If the texture feature is that there is no texture, then the texture classification of the diatom is determined to be empty texture; If the texture feature is a pattern with equal, short, palisade-like structures along the edge, and the palisade-like structures are located on the inside or outside, then the texture classification of the diatom is determined to be a serrated texture; If the texture feature is a pattern with straight, uniform lines at equal intervals throughout the body, forming a striped or linear appearance, then the texture classification of the diatom is determined to be a striped texture; If the texture feature is a pattern with straight, uniform lines running through the entire body at equal intervals, forming a striped or linear appearance, then the texture classification of the diatom is determined to be a vertical line texture; If the texture feature is a pattern with a prominent single vertical line running through the entire body from top to bottom and located in the center, the texture classification of the diatom is determined to be a cross-shaped texture; If the texture feature is a pattern defined by evenly distributed horizontal lines intersecting a single vertical line running from top to bottom, then the texture classification of the diatom is determined to be a palisade texture; If the texture feature is a pattern consisting of two concentric circles, and the edge of each circle may be decorated with a radial pattern, then the texture classification of the diatom is determined to be a concentric circle texture.

9. The diatom morphological classification method for forensic drowning location inference according to claim 1, characterized in that: The method of respectively extracting three principal component variables from the diatom vector data and the tissue vector data includes respectively extracting the first three principal component variables from the diatom vector data and the tissue vector data based on principal component analysis.

10. The diatom morphological classification method for forensic drowning location inference according to claim 1, characterized in that: The method of calculating the distance between the first coordinate point and the second coordinate point corresponding to each sampling point includes cosine distance or Euclidean distance.

Images