19 results about "Seed testing" patented technology

A diversified validation test suite application (DVTSA) for a generative AI powered tool or large language model (LLM) tool includes at least first, second, and third control logics. The first control logic receives a seed test input or user input, analyzes the seed test input or user input, and extracts key elements for variations. The second control logic performs a coverage measurement of outputs of the first control logic. The third control logic causes a human validator to evaluate outputs of the second control logic relative to predefined coverage metrics and selectively and continuously iterate to cause outputs of the second control logic to increase LLM tool input robustness and output robustness from a first level to a second level greater than the first level, in both production and pre-production LLM tool processes.

This invention discloses an IoT fuzzing method based on LLM-guided and FSM dynamic inference, belonging to the field of IoT network security and software testing technology. Addressing the problems of low coverage and inaccurate state machine inference in current IoT protocol fuzzing, this invention first constructs an initial FSM by combining IoT protocol specifications and captured traffic data. Then, it generates a large number of test cases through mutation of seed test cases for fuzzing testing. Features are extracted from device responses, and state identification is performed by calculating similarity. When a new state appears, the FSM and state fingerprint database are updated. When coverage becomes a bottleneck, LLM-guided path inference is used to generate extended sub-FSMs and test cases, which are then executed. The FSM is then corrected based on the test results. This invention enables high-precision automated construction of IoT protocol state FSMs, improving test coverage and enhancing the efficiency and accuracy of vulnerability discovery.

1. Name of the product in this design: Rice Seed Testing Machine. 2. Purpose of this design: for high-precision intelligent seed testing of single rice panicles. 3. The key design feature of this product is its shape. 4. The image or photograph that best illustrates the design's key features: the front view.

The invention relates to the technical field of seed testing equipment, in particular to a beet pelletized seed wear resistance testing device, which comprises a roller provided with a testing cavity; the driving assembly is used for driving the roller to rotate; the collision assembly comprises a protruding mechanism and a fixing mechanism; the cleaning assembly comprises an impact gas storage mechanism, an energy storage mechanism, a storage mechanism, an injection mechanism and a dust collection mechanism; according to the collision assembly, through direct collision between the protruding mechanism and the pelletized seeds, the requirement for testing the wear resistance of the seeds is accurately simulated; through linkage of the collision air storage mechanism and the injection mechanism, seed collision kinetic energy is reasonably utilized, collision energy is recycled and converted into compressed air and electric energy, the stored energy can be utilized to clean the inner wall of the roller through the cleaning assembly after a test is finished, energy recycling is achieved, and the subsequent test precision is effectively improved.

This invention relates to the field of soybean seed testing technology, and more particularly to a device for soybean seed testing. The device includes a pod feeding port, a pod shell discharge port, and a soybean discharge port located outside the left side of the main body of the device. Inside, there are pod feeding, imaging, soybean shelling, and soybean imaging units, enabling pod conveying, pod image acquisition, shelling separation, and soybean image acquisition. A straw scanning chamber is located outside the right side of the main body of the device, containing a Y-axis motion unit, an electrical control unit, an electric turntable, a chuck, and a 3D scanning unit. The electric turntable and 3D scanning unit can scan the straw in all directions. The device also includes control buttons and a display for operation and data display. This technical solution solves the problems of low efficiency, subjective data, and lack of comprehensive analysis of key traits of pods and straw in manual seed testing, making it impossible to achieve a comprehensive, efficient, and accurate evaluation of soybean traits.

The utility model relates to the technical field of seed testing devices, and discloses a tetraploid rice seed testing device which comprises a machine body and a feeding cylinder fixedly connected to the machine body, the top of the feeding cylinder is fixedly connected with a feeding hopper, the top of the inner wall of the feeding cylinder is slidably connected with a vibration frame, one side of the feeding cylinder is provided with a through groove matched with the vibration frame, and the through groove is communicated with the feeding hopper. A vibrating assembly for vibrating the vibrating frame is fixedly connected to one side of the feeding cylinder and located below the through groove, and a mounting groove is formed in the top of the vibrating frame; according to the tetraploid rice seed testing device, after rice sample seeds fall onto the filter plate through the feeding hopper, the filter plate can vibrate through the operation of the vibration assembly, so that the screening efficiency of the filter plate on the rice sample seeds is improved, impurities can be discharged through the discharging groove after the rice sample seeds are screened by the filter plate, and the screening efficiency is improved. A worker does not need to frequently clean impurities on the filter plate, so that the device is more convenient to use, and the rice seed testing efficiency of the device is effectively improved.

The application discloses a kind of fuzzy testing methods for continuous integration process, it is related to software collaborative development field.The application includes the following steps: in the project of applying continuous integration, difference analysis is carried out to two adjacent submissions, the difference information of both is obtained, and change point is obtained according to difference information, and is stored in change point set;In the process of building project, data flow analysis is carried out according to change point information, and data flow analysis result is obtained;Program is inserted into using data flow analysis result, and the measured program that has been inserted is obtained;The measured program is carried out fuzzy testing, the fitness of seed test case is calculated in the testing process, and corresponding test resource is allocated to it according to fitness.The application more specifically tests the place where change is generated in continuous integration process, and reduces the overhead brought by manual construction test case.

The present invention relates to the technical field of soybean seed testing, and in particular to a soybean seed testing system and method improved based on YOLO v8. The system comprises a pod recognition network and a pod image restoration network. The pod recognition network adopts an improved YOLO v8 network combined with a hybrid attention module in a detection head to identify pods and stems, thereby obtaining a stem image in a pod image and an independent image of each pod in the pod image. The pod image restoration network adopts an adversarial neural network to restore the independent pod image. The present invention introduces an attention mechanism for extracting local and global features, so that a target frame is more accurately positioned in a target area, thereby improving detection accuracy. The method uses a loss function combined with ShapeIoU loss, calculates the loss by focusing on the shape and scale of the bounding box itself, thereby making the bounding box regression more accurate, and combines the calculation and statistics of the characteristics of the independent pod image to complete high-precision soybean seed testing.

A diversified validation test suite application (DVTSA) for a generative artificial intelligence (AI) driven tool or a large language model (LLM) tool includes at least a first control logic, a second control logic, and a third control logic. The first control logic receives a seed test input or a user input, analyzes the seed test input or the user input, and extracts varying key elements. The second control logic performs an overlay measurement of the output of the first control logic. The third control logic causes the human verifier to evaluate an output of the second control logic with respect to a predefined coverage metric, and selectively and continuously iterating such that the output of the second control logic increases LLM tool input robustness and output robustness from a first level to a second level greater than the first level during production and pre-production LLM tools.

The utility model aims to provide a water-replenishing-free simple germination box device for seed inspection, and relates to the technical field of seed inspection, a water storage tank is rotatably arranged on a base, a plurality of germination boxes are arranged on the outer circumference of the water storage tank in an array mode, and outer boxes of the germination boxes are communicated with the water storage tank through upper pipelines and lower pipelines; water in the water storage tank is supplemented into the outer tank through the lower pipeline, and the liquid level in the outer tank is kept at a certain height through the upper pipeline, so that the sponge in the inner tank is always soaked in the water, the humidity of seeds in the inner tank is kept, regular manual watering is not needed, and the workload is reduced; the moving part is arranged on the outer box in a sliding mode, during use, the distance between the inner box and the bottom of the outer box is controlled by sliding the moving part, the soaking depth of sponge at the bottom of the inner box in water is controlled, and therefore the humidity of seeds is controlled; the sliding of the moving part is limited through the cooperation of the limiting columns and the limiting holes in the outer box.

This invention discloses a method for correcting seed testing data of *Gnaphalium affine* based on transfer learning, belonging to the field of seed quality testing technology. This method collects benchmark data from standard equipment under standard conditions and target data from different testing equipment and environmental conditions to construct an equipment-environment error database and fit an error influence function. The benchmark data is set as the source domain, and the target data as the target domain. A transfer correction model is trained to correct the target domain data, and the classification threshold is dynamically updated. The corrected data is then mapped to a unified feature space to verify consistency, ultimately outputting standardized results and data traceability information applicable across laboratories. This method effectively eliminates systematic errors, standardizes testing data, improves correction accuracy and data reusability, ensures result reliability and traceability, and provides technical support for the standardization of seed testing.

This invention discloses an intelligent rice variety classification system and method based on image recognition, aiming to address the technical pain points of existing rice variety classification methods, such as reliance on manual experience, low efficiency, large errors, incomplete feature extraction, insufficient classification accuracy, and poor adaptability. The system includes a dual-optical-path controllable image acquisition module, a rice-specific preprocessing module, a dual-branch feature decoupling and fusion extraction module, a few-sample classification inference module, and a result output and verification module. The classification method sequentially performs image acquisition, preprocessing, feature extraction, variety identification, and result verification. Dual-path images of rice varieties are acquired through dual optical paths, and after preprocessing, the dual-branch module extracts fused features, which are then identified by the few-sample classification module. This invention achieves non-destructive, rapid, and accurate intelligent classification of rice varieties, improves the accuracy of distinguishing highly similar rice varieties, reduces usage and maintenance costs, adapts to multiple scenarios, and is applicable to fields such as agricultural seed testing and breeding research.

A diversified validation test suite application (DVTSA) for a generative AI-powered tool or a large language model (LLM) includes at least a first, a second, and a third control logic. The first control logic receives seed test input or user input, analyzes the seed test input or user input, and extracts key elements for variations. The second control logic performs a coverage measurement of the outputs of the first control logic. The third control logic prompts a human tester to evaluate the outputs of the second control logic against predefined coverage metrics and to iterate selectively and continuously to ensure that the outputs of the second control logic increase the LLM tool's input and output robustness from a first level to a second level that is higher than the first, in both production and pre-production LLM tool processes.

The present invention relates to the field of soybean seed testing technology, and more particularly to a soybean seed testing system and method based on an improved Faster-R-CNN. In this system, a pod recognition network accurately identifies pods and stems in their naturally mature state, while a pod image restoration network addresses the issue of mutual occlusion between pods. By optimizing the network structure, the efficiency and accuracy of phenotypic recognition in complex scenarios are significantly improved. The method utilizes a recognition loss function combined with Focal_Loss loss to reduce the loss contribution of easily classified samples, allowing the model to focus more on difficult-to-classify samples, thereby helping to address class imbalance and improving model robustness.

The present invention relates to the field of soybean seed testing technology, and in particular to a soybean seed testing system and method based on an improved Transformer-faster-RCNN. In this system, a pod recognition network based on a Swin-Transform structure accurately identifies pods and stems in their naturally mature state. A pod image restoration network with an adversarial neural network structure addresses the problem of mutual occlusion between pods. By optimizing the network structure, the efficiency and accuracy of phenotypic recognition in complex scenarios are significantly improved. The method uses a recognition loss function combined with Focal_Loss loss to reduce the loss contribution of easily classified samples, allowing the model to focus more on difficult-to-classify samples, thereby helping to address the problem of class imbalance and improving the robustness of the model.

This invention relates to the field of soybean seed testing technology, and more particularly to a seed testing device for soybean pods and seeds. The device includes a pod feeding port, a pod shell discharge port, and a soybean discharge port located outside the main body of the device, and a pod feeding unit, a pod imaging unit, a soybean dehulling unit, and a soybean imaging unit located inside the device. This enables pod conveying, pod image acquisition, dehulling separation, and soybean image acquisition. The device also includes control buttons and a display for operation and data display. Through the above technical solution, a fully automated process is achieved from pod feeding, image acquisition, dehulling to soybean seed detection, reducing manual intervention, significantly shortening the seed testing time, and improving seed testing efficiency. It solves the problems of low efficiency and susceptibility to subjective factors in traditional seed testing methods, making it difficult to guarantee the accuracy and consistency of data.

The utility model relates to a sampling device for multi-station seed detection. The sampling device comprises a barrel with a hollow structure, and a sampling assembly is arranged in the cylinder body. The sampling assembly comprises a sampling head close to one end of the barrel for sampling, a sampling rod connected with the sampling head and a pushing piece positioned at one end of the barrel far away from the sampling head. A connecting assembly is arranged on the circumferential side face of the barrel. According to the multi-station sampling device, multiple parts of multi-station samples can be obtained from equidistant positions of a seed pile at the same height through one-time sampling work, and the working efficiency is effectively improved. Moreover, the nesting design of the sampling assembly and the cylinder facilitates safe placement of the sampling device when the sampling device is idle, avoids possible damage caused by direct exposure of the sampling head to the outside, and avoids mutual collision damage among a plurality of sampling devices, so that subsequent multi-station seed sampling work is facilitated.

The application discloses an intelligent peanut seed testing method based on a neural network model, first constructs a seed testing network model and carries out training, tests peanut seeds based on the trained network model, calculates a breakage rate, identifies a seed category, and determines a hundred-grain weight and a length-width ratio of the peanut. The scheme constructs the seed testing network model, when clustering an initial anchor frame, solves the problem that convergence is seriously dependent on initialization of a cluster center, avoids the problem that a local optimal solution is tended to when clustering the initial anchor frame, uses 1-iou instead of Euclidean distance, can obtain an anchor frame with better precision and higher accuracy, a CBAM sub-module is redesigned, a feature extraction network based on Darknet in an original network is removed, deep separable convolution and an inverse residual structure are added, the running speed and accuracy of the neural network are effectively improved, and intelligent seed testing of the peanut is realized.

The invention discloses an automatic soybean seed testing instrument, and relates to the technical field of soybean seed testing, the automatic soybean seed testing instrument comprises a weighing assembly, a threshing assembly is arranged on the left side of the top of the weighing assembly, a stalk crushing assembly is nested in the middle of the right side of the weighing assembly, and a separation counting assembly is arranged on the top of an inner cavity of the weighing assembly; a phenotype data acquisition assembly is arranged at the bottom of the separation counting assembly, and a soybean particle distribution assembly is arranged below the phenotype data acquisition assembly; the soybean particle distribution assembly comprises a fixing frame fixedly arranged at the top of the base, a bearing disc is fixedly arranged at the top of the fixing frame, and a plurality of containing channels are evenly formed in the inner side of the bearing disc. Compared with traditional manual threshing, the threshing efficiency is higher, compared with threshing of traditional threshing equipment, after soybean threshing is completed, the steps of transferring threshed soybean particles and the like are not needed, the automation degree is higher, and the seed testing efficiency can be effectively improved.