An automatic counting method for the egg production or hatching rate of Spodoptera exigua

Abstract

The application discloses an automatic counting method for the egg-laying quantity or hatching rate of Spodoptera exigua. The method comprises the following steps: collecting a multilayer focal plane image sequence of an egg mass, constructing a height field model of the egg mass, dividing egg grains into different physical layers based on the spatial position relationship in the vertical direction, and respectively counting the number of each layer. The method also introduces a shielding compensation calculation to correct the lower layer of the blocked egg grains, so as to accurately obtain the layered egg quantity data. The application solves the problem that the traditional two-dimensional image recognition cannot distinguish the hierarchy of the egg mass, which leads to a serious low count, realizes the accurate counting of the stacked egg mass, and provides a reliable technical means for entomological research and pest forecasting.

Description

Technical Field

[0001] This invention relates to the field of agricultural entomology, and more specifically, to an automatic counting method for the number of eggs laid or the hatching rate of the beet armyworm indoors. Background Technology

[0002] The beet armyworm (Spodoptera litura) is a global agricultural pest with a high reproductive capacity and a wide host range, causing serious damage to various crops. In insect ecology research and pest forecasting, accurately counting the beet armyworm's oviposition and hatching rate is crucial data for assessing its population dynamics and developing control targets. Therefore, in indoor reproductive toxicology experiments of insecticides, oviposition is an important indicator for evaluating the pest's reproductive viability. The beet armyworm exhibits unique oviposition characteristics (small eggs, with a single egg diameter of only 0.5 to 0.6 mm; egg masses arranged in a fish-scale pattern, exhibiting single, double, and multi-layered patterns), and the oviposition location and area are random.

[0003] Traditional egg counting relies primarily on manual visual inspection using a microscope, which is not only labor-intensive and inefficient, but also prone to causing visual fatigue and resulting in missed or duplicate counts. In recent years, with the development of machine vision technology, some automated counting devices have emerged, such as the patent with publication number CN1261866C, which uses image acquisition and threshold segmentation technology to count tiny insects and their eggs. However, existing technologies have significant drawbacks when dealing with Spodoptera litura egg masses: female Spodoptera litura lay eggs in clusters, often forming multi-layered egg masses (eggs stacked on top of each other) rather than a single layer. Traditional two-dimensional image processing techniques can only acquire the top projected area of ​​the egg mass and cannot identify the number of stacked layers of eggs in the vertical direction, leading to severely undercounted results.

[0004] Therefore, the existing technology has defects and urgently needs improvement. Summary of the Invention

[0005] To address at least one of the aforementioned technical problems, the present invention aims to provide an automatic counting method for the number of eggs laid or the hatching rate of the beet armyworm indoors, which can achieve accurate counting of stacked egg masses, providing a reliable technical means for entomological research and pest monitoring.

[0006] This invention provides an automatic counting method for the number of eggs laid or the hatching rate of the beet armyworm indoors, comprising: Obtain a multi-layer focal plane image sequence of Spodoptera litura egg masses; Based on the multi-layer focal plane image sequence, a height field model of the egg mass is constructed; Identify and segment individual eggs in the height field model, and determine the spatial position of the eggs in the vertical direction; Based on the spatial relationship of different eggs in the vertical direction, the identified eggs are divided into different physical layers; The number of eggs in each physical layer was counted to obtain the egg quantity data for each layer. Based on the stratified egg quantity data, combined with a preset time series image, the dynamic changes in egg production or hatching rate are calculated.

[0007] In this scheme, the step of classifying the identified eggs into different physical layers based on their spatial positional relationship in the vertical direction specifically includes: Using the surface of the medium supporting the egg mass as the reference zero plane, the height value of the corresponding egg relative to the reference zero plane is determined according to the spatial position relationship of the egg in the vertical direction. The height range into which the egg falls relative to the reference zero plane: Determine the physical layer n of the current egg; where H represents the preset vertical layer height threshold, n is a natural number greater than or equal to 1, and "*" indicates multiplication.

[0008] This plan also includes: For the nth layer of eggs, obtain the area of ​​occlusion of the (n-1)th layer by its covered area; The number of eggs blocked in the (n-1)th layer is determined based on the blocked area and the preset average cross-sectional area of ​​the eggs. The original count result of (n-1) is corrected based on the number of eggs that are blocked in the (n-1)th layer.

[0009] In this scheme, the formula for correcting the original count result of the (n-1)th layer based on the number of eggs obscured in the (n-1)th layer is as follows: ;in This represents the corrected number of eggs in the (n-1)th layer. Let m represent the projected occlusion area of ​​the i-th egg in the n-th layer on the (n-1)-th layer, and m represent the total number of eggs in the n-th layer, where i ∈ m. This represents the average cross-sectional area of ​​the egg. This represents the number of eggs directly observed in the (n-1)th layer.

[0010] This plan also includes: When there are eggs in the egg mass that are completely obscured by the upper layer of eggs, the number of eggs in the (n-1)th layer that are completely obscured is set as The calculation formula is as follows: Where k represents the calibration factor, This represents the density of the egg cells in the nth layer. This represents the total area covered by the nth layer of eggs. This represents the number of eggs that have been identified in the nth layer. Indicates the total area of ​​the egg mass; The number of eggs in the (n-1)th layer is further adjusted based on the number of eggs completely obscured in the (n-1)th layer to obtain the final number of eggs.

[0011] This plan also includes: Extract the diameter of visible eggs along their minor axis and construct a set of egg diameters; Determine the variance of the corresponding set of egg diameter data based on the set of egg diameters; If the variance is greater than the preset variance threshold, then the current egg is set as having uneven egg size; Based on the current uneven egg size, the formula for calculating the preset vertical layer height threshold is as follows: ,in This represents the average minor axis diameter of the visible eggs. This indicates the preset scaling factor.

[0012] In this scheme, the step of constructing a height field model of the egg mass based on the multi-layer focal plane image sequence specifically includes: Sharpness analysis was performed on the multi-focal plane image sequence to determine the sharpness peak of each egg on different focal planes; Based on the focal plane position information corresponding to the peak resolution, the relative height of each point on the surface of the egg is calculated, thereby generating a height field model of the egg mass.

[0013] In this plan, the formula for calculating the hatching rate is as follows: ,in This indicates the total number of eggs before hatching. This indicates the number of eggs remaining after hatching.

[0014] This plan also includes: Acquire multi-layer images of the egg mass after hatching; Extract the average gray value G of each identified egg cell; The average gray value G is compared with the average gray value of eggs at the same location before hatching. If a comparative analysis is conducted, If the egg is an empty shell that has already hatched, then it is determined that the egg is an empty shell. This indicates the preset grayscale change threshold.

[0015] In this scheme, when the egg mass is located at the boundary of the image acquisition area, it also includes: Based on a preset recognition algorithm, extract the ovals that intersect with the image edges; Extract the effective area of ​​the egg particles intersecting the image edge located inside the image. ; The effective area of ​​the egg grains intersecting the image edge located inside the image is analyzed by comparing it with the preset total area to determine the area ratio r; If the egg is assigned to the nth layer, the count of the nth layer increases by r, instead of 1.

[0016] One or more technical solutions proposed in this application have at least the following technical effects: 1. By constructing a height field model of egg masses, we can overcome the limitations of traditional two-dimensional counting and achieve accurate three-dimensional hierarchical counting; 2. To address the issue of partial occlusion of lower-layer eggs by upper-layer eggs, this invention designs a geometric compensation formula based on the projected area. Furthermore, to address the issue of complete occlusion in extreme stacking scenarios, compensation calculations are performed to achieve a reasonable estimation of multi-layer stacked eggs. 3. To address the difficulty in distinguishing between hatched eggshells and unhatched eggs, this invention introduces a grayscale threshold discrimination mechanism. By comparing the average grayscale value change ΔG of eggs at the same location before and after hatching, and utilizing the difference in optical properties—empty shells becoming brighter and dead eggs becoming darker—high-precision automatic discrimination of the hatching status is achieved. This mechanism avoids subjective errors from manual visual inspection and provides a reliable data foundation for the automated calculation of hatching rates. Attached Figure Description

[0017] Figure 1 A flowchart of an automatic counting method for the number of eggs laid or the hatching rate of the beet armyworm indoors is shown below. Figure 2 A schematic diagram of two layers of eggs in the indoor beet armyworm eggs of the present invention is shown; Figure 3 A schematic diagram of multiple layers of eggs in the indoor beet armyworm eggs of the present invention is shown. Detailed Implementation

[0018] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0019] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0020] Figure 1 The flowchart illustrates an automatic counting method for indoor Spodoptera litura oviposition or hatching rate according to the present invention.

[0021] like Figure 1 As shown, this invention provides an automatic counting method for the number of eggs laid or the hatching rate of the beet armyworm indoors, comprising: S101, Obtain a multi-layer focal plane image sequence of Spodoptera litura egg masses; S102, Based on the multi-layer focal plane image sequence, construct a height field model of the egg mass; S103, Identify and segment individual eggs in the height field model, and determine the spatial position of the eggs in the vertical direction; S104. Based on the spatial relationship of different eggs in the vertical direction, the identified eggs are divided into different physical layers. S105, count the number of eggs in each physical layer to obtain the egg quantity data for each layer; S106. Based on the stratified egg quantity data and combined with a preset time series image, calculate the dynamic changes in egg production or hatching rate.

[0022] According to an embodiment of the present invention, by changing the focal length, such as controlling the up and down movement of the microscope stage or adjusting the liquid lens, multiple photos are continuously taken at different heights to form a multi-layer focal plane image sequence from the bottom to the top of the egg mass. The layered egg quantity data is directly extracted from the multi-layer focal plane image sequence using a preset recognition algorithm (such as Faster R-CNN), and then divided according to different physical layers. In addition, a preset time series image is used to determine the egg production at different time points of the same physical layer. The egg production at the later time point is subtracted from the egg production at the earlier time point to determine the dynamic change of egg production between adjacent time points.

[0023] According to an embodiment of the present invention, the step of classifying the identified eggs into different physical layers based on the spatial positional relationship of different eggs in the vertical direction specifically includes: Using the surface of the medium supporting the egg mass as the reference zero plane, the height value of the corresponding egg relative to the reference zero plane is determined according to the spatial position relationship of the egg in the vertical direction. The height range into which the egg falls relative to the reference zero plane: Determine the physical layer n of the current egg; where H represents the preset vertical layer height threshold, n is a natural number greater than or equal to 1, and "*" indicates multiplication.

[0024] It should be noted that, based on the characteristic that the eggs of the beet armyworm are oblate, the height of the eggs in the vertical direction is approximately equal to the length of their minor axis. Therefore, the preset vertical layer height threshold can be set by calculating the length of the minor axis of the eggs identified in the same egg layer. For example, if the average minor axis diameter of the eggs identified in the second layer is 0.5 mm, then the corresponding layer height threshold can be set to 0.5 mm. Thus, starting from the surface of the medium supporting the egg mass and moving upwards, the eggs with a height between 0 and H are divided into the first layer, those between H and 2H are divided into the second layer, and so on.

[0025] According to an embodiment of the present invention, it further includes: For the nth layer of eggs, obtain the area of ​​occlusion of the (n-1)th layer by its covered area; The number of eggs blocked in the (n-1)th layer is determined based on the blocked area and the preset average cross-sectional area of ​​the eggs. The original count result of (n-1) is corrected based on the number of eggs that are blocked in the (n-1)th layer.

[0026] According to an embodiment of the present invention, the formula for correcting the original count result of the (n-1)th layer based on the number of eggs obscured in the (n-1)th layer is specifically as follows: ;in This represents the corrected number of eggs in the (n-1)th layer. Let m represent the projected occlusion area of ​​the i-th egg in the n-th layer on the (n-1)-th layer, and m represent the total number of eggs in the n-th layer, where i ∈ m. This represents the average cross-sectional area of ​​the egg. This represents the number of eggs directly observed in the (n-1)th layer.

[0027] It should be noted that, assuming there is a larger egg in the second layer, its projected area on the first layer is 0.2 square millimeters. By calculating the average cross-sectional area of ​​the eggs in the entire layer, we know that the average cross-sectional area of ​​the corresponding egg is 0.05 square millimeters. Therefore, the larger egg is likely to completely or entirely cover approximately 0.2 / 0.05 = 4 eggs in the first layer. If 10 eggs are counted by direct observation in the first layer, then the corrected count of eggs in the first layer should be 10 + 4 = 14.

[0028] According to an embodiment of the present invention, it further includes: When there are eggs in the egg mass that are completely obscured by the upper layer of eggs, the number of eggs in the (n-1)th layer that are completely obscured is set as The calculation formula is as follows: Where k represents the calibration factor, This represents the density of the egg cells in the nth layer. This represents the total area covered by the nth layer of eggs. This represents the number of eggs that have been identified in the nth layer. Indicates the total area of ​​the egg mass; The number of eggs in the (n-1)th layer is further adjusted based on the number of eggs completely obscured in the (n-1)th layer to obtain the final number of eggs.

[0029] It should be noted that the calibration factor was obtained through experimental calibration and its value ranges from 0.7 to 0.9; the density of the nth layer of eggs is equal to the total area occupied by the nth layer of eggs divided by the total usable area of ​​the region containing the nth layer, where... The larger the value, the denser the arrangement of the eggs in the nth layer, and the smaller the gaps between the eggs. This means that the occlusion of the next layer is more complete, and the probability of completely occluding the next layer is greater. For example, if the second layer is identified... Eggs, total area of ​​egg mass The area covered by the second layer of eggs This means that the projection of these 50 eggs covers an area of ​​8 square millimeters, and the second layer's arrangement density... If the calibration factor is 0.8, then the number of eggs completely obscured by the first layer is: The number of eggs completely obscured by the first layer is added to the corrected number of eggs in the first layer. This yields the final number of eggs after further correction.

[0030] According to an embodiment of the present invention, it further includes: Extract the diameter of visible eggs along their minor axis and construct a set of egg diameters; Determine the variance of the corresponding set of egg diameter data based on the set of egg diameters; If the variance is greater than the preset variance threshold, then the current egg is set as having uneven egg size; Based on the current uneven egg size, the formula for calculating the preset vertical layer height threshold is as follows: ,in This represents the average minor axis diameter of the visible eggs. This indicates the preset scaling factor.

[0031] It should be noted that eggs laid by the same female insect may vary in size, especially in the early and late stages of oviposition. If the same H is used for all areas, small eggs may be incorrectly classified into higher layers, or large eggs may be squeezed into the same layer. The preset scaling factor ranges from 0.8 to 1.2. The preset scaling factor is used to dynamically adjust the preset vertical layer height threshold to improve the granularity accuracy of the count.

[0032] According to an embodiment of the present invention, the step of constructing a height field model of the egg mass based on the multi-layer focal plane image sequence specifically includes: Sharpness analysis was performed on the multi-focal plane image sequence to determine the sharpness peak of each egg on different focal planes; Based on the focal plane position information corresponding to the peak resolution, the relative height of each point on the surface of the egg is calculated, thereby generating a height field model of the egg mass.

[0033] It's important to note that for each tiny region in the image, the algorithm analyzes its sharpness across different focal planes. The sharper the image, the sharper its edges and the higher its gradient value. The sharpness value of this tiny region is plotted as a curve with the focal plane position as the abscissa. This curve typically has a distinct peak; the focal plane position corresponding to this peak is the actual height of the tiny region. For example, when photographing an egg, if the focus is on the bottom, the bottom edge of the egg is sharp, while the top is blurry; when focused on the middle, the middle is the sharpest; and when focused on the top, the top is the sharpest. The algorithm performs this operation on thousands of points on the egg's surface to obtain the height of each point. Finally, all the height information from these points is pieced together to form the 3D contour of the egg. All the contours of the entire egg mass are then combined to form the height field model of the egg mass.

[0034] According to an embodiment of the present invention, the formula for calculating the hatching rate is as follows: ,in This indicates the total number of eggs before hatching. This indicates the number of eggs remaining after hatching.

[0035] It should be noted that the total number of eggs before hatching is the sum of eggs in all egg layers before hatching; the number of eggs remaining after hatching is the sum of unhatched eggs in all egg layers after hatching.

[0036] According to an embodiment of the present invention, it further includes: Acquire multi-layer images of the egg mass after hatching; Extract the average gray value G of each identified egg cell; The average gray value G is compared with the average gray value of eggs at the same location before hatching. If a comparative analysis is conducted, If the egg is an empty shell that has already hatched, then it is determined that the egg is an empty shell. This indicates the preset grayscale change threshold.

[0037] It should be noted that after insects hatch, the larvae will bite through the eggshell and crawl out, leaving behind an empty shell that is usually transparent, translucent, or lighter in color than a fresh egg; while unhatched, dead eggs may be shriveled and deteriorated inside, and their color will be darker; when the average gray value G of the egg is greater than When the time is right, it indicates that the egg is an empty shell that has already hatched; the preset grayscale change threshold The calculation is based on the known gray values ​​of empty shells and dead eggs. For example, the gray value changes of 10 empty shells and 10 dead eggs before and after hatching are calculated to obtain 20 corresponding gray value changes. Then, the most representative gray value change is extracted from these 20 gray value changes as the corresponding preset gray value change threshold.

[0038] According to an embodiment of the present invention, when the egg mass is located at the boundary of the image acquisition area, the method further includes: Based on a preset recognition algorithm, extract the ovals that intersect with the image edges; Extract the effective area of ​​the egg particles intersecting the image edge located inside the image. ; The effective area of ​​the egg grains intersecting the image edge located inside the image is analyzed by comparing it with the preset total area to determine the area ratio r; If the egg is assigned to the nth layer, the count of the nth layer increases by r, instead of 1.

[0039] It should be noted that during image acquisition, the egg mass cannot always fall completely in the center of the field of view. Some eggs may be cut off by the image boundary, with only a part exposed. If these boundary eggs are simply ignored, the count will be too low. If they are all counted as a complete egg, the count will be too high. Therefore, this invention uses an area ratio counting method to count the boundary eggs of each layer proportionally. For example, if there is an egg on the boundary, the corresponding egg count is r, not counted as 1.

[0040] This invention discloses an automatic counting method for the egg-laying or hatching rate of the beet armyworm indoors. It constructs a height field model of the egg mass by acquiring a multi-layered focal plane image sequence of egg masses. Based on the spatial relationship in the vertical direction, the eggs are divided into different physical layers, and the number of eggs in each layer is counted separately. Furthermore, it introduces occlusion compensation calculations to correct for eggs in the lower layers that are obscured, thereby accurately obtaining layered egg quantity data. This invention solves the problem of traditional two-dimensional image recognition failing to distinguish egg mass layers, leading to severely understated counts. It achieves accurate counting of stacked egg masses, providing a reliable technical means for entomological research and pest monitoring.

[0041] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0042] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0043] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0044] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0045] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

Claims

1. A method for automatically counting the number of eggs laid or the hatching rate of *Spodoptera litura* indoors, characterized in that, include: Obtain a multi-layer focal plane image sequence of Spodoptera litura egg masses; Based on the multi-layer focal plane image sequence, a height field model of the egg mass is constructed; Identify and segment individual eggs in the height field model, and determine the spatial position of the eggs in the vertical direction; Based on the spatial relationship of different eggs in the vertical direction, the identified eggs are divided into different physical layers; The number of eggs in each physical layer was counted to obtain the egg quantity data for each layer. Based on the stratified egg quantity data, combined with a preset time series image, the dynamic changes in egg production or hatching rate are calculated.

2. The method for automatically counting the number of eggs laid or the hatching rate of *Spodoptera litura* indoors according to claim 1, characterized in that, The step of classifying the identified eggs into different physical layers based on their spatial position in the vertical direction specifically includes: Using the surface of the medium supporting the egg mass as the reference zero plane, the height value of the corresponding egg relative to the reference zero plane is determined according to the spatial position relationship of the egg in the vertical direction. The height range into which the egg falls relative to the reference zero plane: Determine the physical layer n of the current egg; where H represents the preset vertical layer height threshold, n is a natural number greater than or equal to 1, and "*" represents multiplication.

3. The method for automatically counting the number of eggs laid or the hatching rate of *Spodoptera litura* indoors according to claim 2, characterized in that, Also includes: For the nth layer of eggs, obtain the area of ​​occlusion of the (n-1)th layer by its covered area; The number of eggs blocked in the (n-1)th layer is determined based on the blocked area and the preset average cross-sectional area of ​​the eggs. The original count result of (n-1) is corrected based on the number of eggs that are blocked in the (n-1)th layer.

4. The method for automatically counting the number of eggs laid or the hatching rate of *Spodoptera litura* indoors according to claim 3, characterized in that, The formula for correcting the original count result of the (n-1)th layer based on the number of eggs obscured in the (n-1)th layer is as follows: ;in This represents the corrected number of eggs in the (n-1)th layer. Let m represent the projected occlusion area of ​​the i-th egg in the n-th layer on the (n-1)-th layer, and m represent the total number of eggs in the n-th layer, where i ∈ m. This represents the average cross-sectional area of ​​the egg. This represents the number of eggs directly observed in the (n-1)th layer.

5. The method for automatically counting the number of eggs laid or the hatching rate of *Spodoptera litura* indoors according to claim 4, characterized in that, Also includes: When there are eggs in the egg mass that are completely obscured by the upper layer of eggs, the number of eggs in the (n-1)th layer that are completely obscured is set as The calculation formula is as follows: Where k represents the calibration factor, This represents the density of the egg cells in the nth layer. This represents the total area covered by the nth layer of eggs. This represents the number of eggs that have been identified in the nth layer. Indicates the total area of ​​the egg mass; The number of eggs in the (n-1)th layer is further adjusted based on the number of eggs completely obscured in the (n-1)th layer to obtain the final number of eggs.

6. The method for automatically counting the number of eggs laid or the hatching rate of *Spodoptera litura* indoors according to claim 2, characterized in that, Also includes: Extract the diameter of visible eggs along their minor axis and construct a set of egg diameters; Determine the variance of the corresponding set of egg diameter data based on the set of egg diameters; If the variance is greater than the preset variance threshold, then the current egg is set as having uneven egg size; Based on the current uneven egg size, the formula for calculating the preset vertical layer height threshold is as follows: ,in This represents the average minor axis diameter of the visible eggs. This indicates the preset scaling factor.

7. The method for automatically counting the number of eggs laid or the hatching rate of *Spodoptera litura* indoors according to claim 1, characterized in that, The step of constructing a height field model of the egg mass based on the multi-layer focal plane image sequence specifically includes: Sharpness analysis was performed on the multi-focal plane image sequence to determine the sharpness peak of each egg on different focal planes; Based on the focal plane position information corresponding to the peak resolution, the relative height of each point on the surface of the egg is calculated, thereby generating a height field model of the egg mass.

8. The method for automatically counting the number of eggs laid or the hatching rate of *Spodoptera litura* indoors according to claim 1, characterized in that, The formula for calculating the hatching rate is as follows: ,in This indicates the total number of eggs before hatching. This indicates the number of eggs remaining after hatching.

9. The method for automatically counting the number of eggs laid or the hatching rate of *Spodoptera litura* indoors according to claim 8, characterized in that, Also includes: Acquire multi-layer images of the egg mass after hatching; Extract the average gray value G of each identified egg cell; The average gray value G is compared with the average gray value of eggs at the same location before hatching. If a comparative analysis is conducted, If the egg is an empty shell that has already hatched, then it is determined that the egg is an empty shell. This indicates the preset grayscale change threshold.

10. The method for automatically counting the number of eggs laid or the hatching rate of *Spodoptera litura* indoors according to claim 1, characterized in that, When the egg mass is located at the boundary of the image acquisition area, it also includes: Based on a preset recognition algorithm, extract the ovals that intersect with the image edges; Extract the effective area of ​​the egg particles intersecting the image edge located inside the image. ; The effective area of ​​the egg grains intersecting the image edge located inside the image is analyzed by comparing it with the preset total area to determine the area ratio r; If the egg is assigned to the nth layer, the count of the nth layer increases by r, instead of 1.

Images