756 results about "Field detection" patented technology

The invention discloses a network protocol reverse analysis method based on deep learning and a graph neural network, and the method comprises the steps: basic field detection: carrying out the byte-level feature extraction of a binary data stream through sliding window embedding, bidirectional LSTM coding and knowledge enhanced CRF decoding, and outputting a structured field labeling sequence; performing protocol format clustering based on the sequence: calculating a multi-dimensional similarity through an improved Needleman-Wunsch algorithm, realizing automatic classification of unknown protocols in combination with a dynamic density clustering algorithm optimized by LSH, and outputting a protocol cluster; and performing composite structure analysis based on the protocol cluster: constructing a protocol syntax tree based on a graph neural network, performing multiple rounds of message passing through a graph attention network GAT attention mechanism, identifying a nested structure and performing recursive analysis, and generating a multi-level protocol syntax tree. According to the method, the automation degree and accuracy of complex protocol analysis can be remarkably improved.

A real-time quantitative characterization method, equipment and a medium for rock mass evolution are disclosed. The method includes: the multiphase field detection and monitoring data is fused; a deep learning model driven by physical principles is established based on the physical principles of fluid density; the deep learning model is trained using real-time fused data as input and corresponding evolution distribution images as output; a trained deep learning model is used to obtain an evolution distribution image based on multi-phase field detection and monitoring data of different time periods and types; a mathematical model is used to quantitatively characterize of physical and mechanical parameters in the whole process of progressive failure of the dynamic evolution of the rock mass based on the macroscopic mechanical parameters and evolution distribution images synchronized with multiphase field detection and monitoring data.

The invention belongs to the technical field of tunnel safety monitoring, and particularly relates to a rock stratum stress-deformation coupled tunnel safety real-time dynamic modeling method, which comprises the following steps of: embedding a rock mass mechanical relationship into a neural network model as a hard constraint, and establishing explicit mapping of monitoring data and a physical field, a model parameter field is dynamically optimized by adopting ensemble Kalman filtering real-time general field detection data, and a data driving and physical mechanism collaborative deduction mechanism is designed, so that long-term prediction error accumulation is effectively inhibited; a stress field, a deformation field and other physical quantities deduced and output by the model are directly utilized to calculate disaster risk indexes with clear mechanical significance, and a physically interpretable early warning decision is realized by fusing fuzzy reasoning and multi-source risks; the technical problems of physical misalignment, weak long-term generalization and poor early warning interpretability of a data-driven model are solved.

The invention discloses an intelligent translation calibration method based on an RAG knowledge base, and belongs to the technical field of artificial intelligence and natural language processing. In order to solve the problems that an existing large language model (LLM) is insufficient in professional term translation accuracy in text translation, low in RAG system retrieval efficiency and the like, an intelligent calibration triggering mechanism is constructed, so that the LLM can autonomously judge whether translation content needs knowledge base enhancement or not in the modes of field detection, term density analysis, user-defined triggering and the like; a full-amount retrieval strategy is changed, and the system efficiency is improved; an RAG knowledge base retrieval framework based on function calling is constructed, multiple types of knowledge sources such as a translation memory library, a term table and a professional corpus are integrated, and accurate domain knowledge is provided for translation calibration; flexible multi-knowledge-base configuration and priority management are supported, and a user can select knowledge base types and set retrieval strategies and fusion weights according to needs; through a progressive translation optimization process of initial translation, intelligent judgment, knowledge base retrieval and calibration optimization, accurate calibration of different granularities such as term level, sentence level and paragraph level is realized, and a standardized knowledge base interface specification is provided. According to the method, the defects of LLM in the aspects of translation accuracy and knowledge base utilization efficiency in the professional field are effectively overcome, and the accuracy and efficiency of knowledge-intensive translation tasks are remarkably improved.

The invention relates to the technical field of soil heavy metal pollution detection, and discloses a soil heavy metal pollution detection method and system, and the method comprises the steps: collecting an original signal containing a target heavy metal characteristic signal and a plurality of auxiliary background signals; preprocessing the original signal to obtain a preprocessed multi-dimensional signal; analyzing the multi-dimensional signal to identify interference characteristics of the target heavy metal characteristic signal and the auxiliary background signal; according to the interference characteristics of the auxiliary background signal, evaluating the composition and intensity of background interference, and according to the composition and intensity of the background interference, separating a pure signal of the target heavy metal from the background by using a nonlinear compensation algorithm; the target heavy metal pure signal is converted into the concentration value of the target heavy metal, chemical digestion is not needed, complex background interference can be effectively recognized and compensated for for soil heavy metal pollution detection, the accuracy and reliability of field detection are remarkably improved, and errors caused by inaccurate reagent degradation and background deduction in a traditional method are overcome.

The invention belongs to the technical field of image processing, and particularly relates to a graphite ore grade detection method based on an improved YOLO11 model, C3k2-CAS and Detect-SEAM modules are introduced, the feature extraction capability is enhanced, the expression of different scales of ore textures is optimized, and the perception capability of the model for grade difference is improved, so that the detection precision is remarkably improved; the attention mechanism of the CAS module is adopted to replace the traditional multiplication operation, the calculation complexity and the model parameter quantity are greatly reduced while the detection performance is maintained, and the improved model is more efficient than a baseline model and is more suitable for edge device deployment; a plurality of data enhancement strategies are combined, so that the model can maintain high robustness and stable detection capability in a complex industrial environment; the trained optimization model can be efficiently deployed to graphite ore grade detection intelligent equipment, real-time and accurate industrial field detection is achieved, the mineral separation efficiency is greatly improved, and an efficient and reliable computer vision solution is provided for the intelligent mining industry.

The invention provides an improved YOLOv11 model-based tea plant disease and insect pest detection method in a complex environment, which comprises the following steps of: acquiring an original image through field shooting, performing data amplification by using a data enhancement technology, and realizing bounding box labeling by using a labeling tool label Img to form a training data set; the method comprises the following steps: constructing a network model of tea plant diseases and insect pests, improving an original model structure of YOLOv11n, and improving a Head module by using an MSDAF module; improving a loss function; performing end-to-end training on the network architecture by using the training set and the test set, and adjusting hyper-parameters; and performing field detection on the tea leaf scab area by using the trained network model. The invention provides a novel tea disease and insect pest detection algorithm based on a YOLOv11 target detection algorithm, and the average precision of tea disease and insect pest detection in a complex environment can be effectively improved.

The invention relates to the technical field of engineering plastic detection, and particularly discloses an aging detection and evaluation method and system suitable for nylon 6 in a marine environment, the system comprises nylon 6 as a core layer, an aging response color-changing material as a surface layer and a middle layer located between the core layer and the surface layer; the thickness of the middle layer is 200-250 [mu] m, and the composite conductive film is transferred to the position 0.2 mm below the surface of the nylon 6 component in a hot pressing mode; the thickness of the surface layer is 0-200 microns and is realized by spraying an aging response color-changing material solution. By integrating the specially designed aging response color-changing material and the conductive film into the nylon 6 matrix, real-time, visual and efficient detection of the aging process is realized. According to the method, the problem that equipment such as a traditional optical fiber spectrometer is complex and expensive is solved, the convenience of field detection is greatly improved through double signal output of color gradient indication and simple electronic reading, and a brand new technical solution is provided for life prediction and preventive maintenance of nylon 6 components in ocean engineering.

An optical detection device, comprising an objective lens, a bright field illumination unit and a dark field illumination unit. The objective lens is provided with a bright field light channel and a dark field light channel. The bright field illumination unit comprises a bright field beam splitting unit and a bright field light source for providing a bright field incident light beam, wherein the bright field incident light beam is reflected by the bright field beam splitting unit and then enters the bright field light channel. The dark field illumination unit comprises a dark field light source for emitting a dark field incident light beam, a prism module and a dark field beam splitting unit; the prism module and the dark field beam splitting unit are sequentially arranged in the transmission direction of the dark field incident light beam; the prism module is used for converting the dark field incident light beam into a circular light beam, wherein the circular light beam is reflected by the dark field beam splitting unit and then enters the dark field light channel; and the dark field beam splitting unit comprises a circular beam splitter and an aperture formed at the center of the circular beam splitter, wherein the aperture is located on the transmission path of the bright field incident light beam. The present invention more conveniently realizes switching between bright field detection and dark field detection, thereby improving the detection efficiency.

The invention relates to the technical field of microelectronics, and discloses a small and special electric machine armature defect detection method based on improved YOLOv11n, which comprises the following steps: S1, acquiring small and special electric machine armature defect image data; s2, constructing an armature defect network model, wherein an original YOLOv11n model structure is improved by constructing a C3K2-MA module; a lightweight feature extraction module is used to replace a convolution module in a trunk feature extraction and down-sampling process, and a down-sampling module is improved; the CIoU of the original YOLOv11n model is improved by using DIOU, so that the model precision is improved, and a loss function is improved; s3, using the training set and the test set obtained in the step S1 to train through the armature defect network model constructed in the step S2; and S4, using the trained armature defect network model to carry out field detection on the armature defects of the small and special electric machine. According to the invention, a partial multi-scale feature aggregation module C3K2-MA is provided, the detection capability of different scale targets is significantly improved, and the detection precision is improved.

The invention provides a chemical fiber cloth defect detection device, and belongs to the field of cloth detection.The chemical fiber cloth defect detection device comprises a bottom plate, two first supporting rods, a placement roller, a fourth motor and a storage roller, a sweeping assembly, a tension adjusting assembly and a detection assembly are installed on the bottom plate, and the detection assembly comprises a detection box fixedly connected to the upper end face of the bottom plate; and a first supporting plate is fixedly mounted on the inner wall of the detection box. The upper and lower irradiation lamps are arranged in the detection box, so that the irradiation effect is better, the device performs bright field detection and dark field detection according to requirements, meanwhile, the angles and heights of the two irradiation lamps can be synchronously adjusted, illumination optimization is realized, distortion and light reflection can be reduced, the comprehensiveness and accuracy of cloth detection are remarkably improved, and the detection efficiency is improved. The light spot interference of a high-reflection area can be avoided, the imaging uniformity is improved, meanwhile, the brush bristles are used for cleaning the surface of the cloth, misjudgment caused by impurities on the cloth during detection is avoided, and the detection accuracy is improved.

The invention relates to the technical field of semiconductor detection, in particular to a wafer bright field illumination and imaging system, a wafer defect detection device and a wafer defect detection method, the system comprises an illumination module, a beam combining module, an objective lens module and an acquisition imaging module; wherein the illumination module generates an illumination light beam adaptive to a wafer detection view field; the beam combining module realizes non-interference separated transmission of an illumination light path and an acquisition light path through a reflection area and a transmission area which work independently; the objective lens module converges the light beams reflected by the reflection area to the surface of the wafer, and controls the light beams to form illumination only at the edge of an entrance pupil so as to realize off-axis illumination; meanwhile, the reflected light beams on the surface of the wafer are collimated and then transmitted to the transmission area; and the acquisition imaging module receives the parallel light beams, converts optical signals into electric signals, and generates a wafer surface image. According to the invention, the problems of high energy loss, high cost, low detection rate and insufficient resolution of a traditional bright field detection system are effectively solved.

The invention provides a light source multiplexing diamond NV color center wide field detection system and an illumination detection method. The system comprises a diamond containing an NV color center, an exciting light module, a first double-color sheet, an objective lens, a second double-color sheet, a wide field imaging module, an illumination imaging module and a microwave module, exciting light generated by the exciting light module irradiates the first double-color piece, is guided to the objective lens, and irradiates the diamond after being transmitted by the objective lens so as to excite the NV color center of the diamond to generate fluorescent light; the objective lens is of an infinity type; the exciting light is irradiated to a sample to be detected after being transmitted by the diamond; fluorescent light and exciting light from a sample to be detected are collected by the objective lens, then transmitted to the first double-color piece and guided to the second double-color piece through the first double-color piece, the second double-color piece guides the fluorescent light to the wide-field imaging module for imaging, and the exciting light in the second double-color piece is guided to the illumination imaging module for imaging. Illumination can be achieved in the exciting light working state, exciting light switching operation in the prior art is avoided, and efficiency can be remarkably improved.

The invention relates to an electrically-assisted biaxial stretching forming limit testing system and method for a plated plate, and belongs to the technical field of bipolar plate testing.The testing method comprises the following steps that a bipolar plate sample is heated through pulse current and then stretched; and meanwhile, temperature, displacement, load and surface deformation images of the bipolar plate sample during stretching are collected, ultrasonic detection is carried out, and a basic forming limit curve is established based on a plastic instability criterion and is dynamically corrected. Dynamic evolution of necking initiation, expansion and interface failure in the stretching process can be synchronously tracked, actual working condition requirements can be better met, interface thermal mismatch aggravated by Joule heat is avoided, and the influence of insufficient sensitivity of a general ultrasonic algorithm can be made up through multi-physics field detection. The failure criterion is redefined as the critical state of plating-substrate interface separation, and the error of the critical failure point predicted by the forming limit curve is further reduced.

The invention discloses a vehicle-mounted defrosting and demisting intelligent control system and method for zoning dynamic adjustment, and relates to the technical field of vehicle-mounted control, and the method comprises the steps: initializing and activating a dual-camera visual sensor array, an infrared temperature field detection device, a pipeline pressure sensor, a vehicle cabin temperature and humidity sensor and a zoning air port execution mechanism; the method comprises the following steps of: performing multi-modal fusion processing on visible light images and infrared temperature data of the double cameras, obtaining characteristic indexes of moisture condensation and ice crystals on the surface of partitioned glass, generating continuous and adjustable total wind speed signals, dynamically distributing the total wind speed signals in a normalized manner according to a preset distribution index, obtaining partitioned wind speed signals, constructing a time sequence prediction model, and predicting the wind speed of the partitioned glass according to the time sequence prediction model. And the atomization risk score is output, a thermal field compensation coefficient is generated through mapping, the thermal field compensation coefficient is issued to a partition air opening executing mechanism, and partition defrosting and demisting are completed. The defrosting and demisting precision, the energy utilization rate and the response speed are remarkably improved by recognizing the frosting and dewing states of the partitions, optimizing the energy efficiency through wind speed distribution and conducting thermal field pre-compensation.

The invention relates to a macro / micro wafer defect detection device, which comprises a macro detection unit, which comprises a macro adjustment assembly, a macro clamping assembly, a light source and a macro camera, and is characterized in that the macro adjustment assembly comprises an R-axis driving mechanism and an F-axis driving mechanism fixedly mounted at the driving end of the R-axis driving mechanism; the macroscopic clamping assembly is fixedly mounted at the driving end of the F-axis driving mechanism, and the macroscopic clamping assembly is used for clamping a wafer; the microcosmic detection unit comprises a microcosmic platform fixedly installed, a first linear driving mechanism fixedly installed on the microcosmic platform, a wafer adsorption mechanism fixed to the driving end of the first linear driving mechanism and a bright and dark field detection mechanism fixedly installed on the microcosmic platform, and the driving direction of the first linear driving mechanism is the horizontal direction; the wafer adsorption mechanism is used for adsorbing a wafer, and the bright and dark field detection mechanism is used for shooting the front and back of the wafer. According to the invention, macroscopic and microscopic detection can be carried out on the surface of the wafer, and the detection efficiency is relatively improved.

The invention provides a GIS equipment partial discharge type classification method based on partial discharge time domain signals, and belongs to the field of partial discharge detection and classification. The problem that the method for judging the partial discharge type through the partial discharge mode spectrogram fails due to the fact that an externally applied voltage signal cannot be obtained in the field detection process is solved. The method comprises the following steps: respectively acquiring time-domain data sets of partial discharge light of a sample in the occurrence process of three partial discharge phenomena of corona discharge, surface discharge and suspension discharge through an experimental platform, and preprocessing the time-domain data sets; adopting a dual-threshold peak searching method to perform dual-threshold peak searching on data in the time domain data sets of the three types of preprocessed partial discharge light to obtain corresponding partial discharge peak data sets, and processing the partial discharge peak data sets into corresponding partial discharge light signal amplitude difference spectrograms; partial discharge type classification is realized through the partial discharge optical signal amplitude difference spectrogram; the method is applied to partial discharge detection of the GIS equipment.

The invention discloses a wheat single-ear gibberellic disease infection rate detection method in the technical field of intelligent agricultural information, and the method comprises the following steps: 1, optimizing a trunk feature extraction network in an AM-YOLOv7 network model; step 2, an activation function in the AM-YOLOv7 network model is optimized; step 3, a feature fusion pyramid network in the AM-YOLOv7 network model is optimized; step 4, the lightweight effect of the optimized AM-YOLOv7 network model is verified; step 5, carrying out transfer learning and model training on the optimized AM-YOLOv7 network model; and step 6, using the optimized AM-YOLOv7 network model to output a detection result. The method ensures that breeders perform real-time detection on the wheat single-ear scab infection rate in an actual scene, is suitable for edge equipment deployment, and solves the problem that the field detection requirement of the wheat single-ear scab infection rate in an actual disease-resistant breeding scene cannot be met due to deep network model hierarchy, large parameter quantity and high requirements on operation memory and computing power.

The invention discloses a lossless detection mold and method for a rammed earth wall, and relates to the technical field of civil engineering and cultural relic protection. In order to solve the problem that in a traditional rammed earth wall nondestructive detection method, it is difficult to establish an effective corresponding relation between signal features and physical and mechanical parameters, the rammed earth wall nondestructive detection mold and method comprise the steps that rammed earth materials are prepared with reference to the soil texture of an ancient wall, and the rammed earth materials are filled into the rammed earth mold in a layered mode to prepare a wall-like test piece; carrying out nondestructive detection on the city-wall-like test piece by adopting a geological radar to obtain an oscillogram and a grey-scale map, and extracting a detection signal; sampling the city-wall-like test piece, carrying out a physical and mechanical test, and establishing a parameter database in which detection signal parameters are associated with physical and mechanical parameters; and finally, carrying out field detection on the ancient city wall, carrying out geological radar detection to obtain an actual detection signal, and matching the actual detection signal with the parameter database to obtain physical and mechanical parameter information of the ancient city wall. The method can effectively improve the accuracy of lossless detection of the rammed earth structure, and provides a scientific basis for ancient city wall protection.

The invention discloses a deep underground multi-physical field detection method and system, and particularly relates to the technical field of underground detection, and the method comprises the following steps: S1, obtaining seismic wave field data; s2, based on a multi-physics field original data set; s3, according to the basic characteristic parameter set; s4, calling a comprehensive feature fusion matrix; in the deep underground multi-physics field detection process, waveform, wave velocity and amplitude parameters contained in seismic wave field data and electromagnetic field intensity, resistivity and polarizability parameters contained in electromagnetic field data are collected by adopting a monitoring instrument sensor to generate a multi-physics field original data set, so that comprehensive basic data are provided for subsequent processing; based on the data set, the wave velocity and amplitude characteristics of each waveform segment of seismic wave field data and the resistivity and polarizability characteristics of electromagnetic field data are determined, a basic characteristic parameter set is obtained through integration, and key characteristics of different physical fields are accurately grasped.

The invention discloses a lateral flow chromatography test strip for detecting perfluorooctanoic acid by combining gold nanoparticles, a nucleic acid aptamer and an antibody. According to the test strip, a nucleic acid aptamer marked by gold nanoparticles is combined with PFOA molecules to form a nucleic acid aptamer with a G-quadruplex structure, and then the nucleic acid aptamer with the G-quadruplex structure is specifically recognized and captured by an anti-BG4 antibody, so that specific visual detection of perfluorooctanoic acid is realized. After a sample is injected, a G-quadruplex structure formed by the PFOA and the nucleic acid aptamer acts with the anti-BG4 antibody coated on the test strip, so that the gold nanoparticles are agglomerated to generate a chromogenic reaction, the signal intensity can be detected by naked eyes or reading equipment, the visual detection limit is 10 mu M, and the equipment detection limit is 304 nM. The test strip has the advantages of high detection speed, high specificity and high sensitivity, and is suitable for field detection and rapid screening. Compared with a traditional method, the method realizes simple, convenient and economical detection of PFOA, and has a wide application prospect.

The invention relates to the technical field of electric field detection and early warning. The invention aims to overcome the defect of lack of an electric field detection device under the hot-line working condition of an ultra-high-voltage line in the prior art, and relates to an electric field detection device under the hot-line working condition of the ultra-high-voltage line, which is a non-contact electric field detection device and is used for detecting the electric field under the hot-line working condition of the ultra-high-voltage line in the hot-line working condition of the ultra-high-voltage line. The non-contact electric field detection device takes a microprocessor as a core control unit, and is composed of a double-spherical-shell type electric field sensor, a differential amplifier, a filter circuit, a true RMS conversion circuit, a power supply module and a man-machine interaction circuit. And deep association of data support, function collaboration and scene adaptation is carried out through the structure. The problem that traditional equipment is insufficient in stability in a severe environment is solved, continuous and stable operation of the electric field detection equipment under complex working conditions is achieved, it is ensured that reliable electric field intensity data can be continuously output, and a basic guarantee is provided for operation safety evaluation.

The invention discloses a surface antigen gene StSAG1 of Sarcocystis tenella, the nucleotide sequence of the surface antigen gene StSAG1 is as shown in SEQ ID NO: 1, and according to the analysis result of protein antigenicity and antigen epitope, the 686 bp-1175 bp segment of the StSAG1 gene is selected to construct a recombinant protein expression vector and engineering bacteria; the antigen disclosed by the invention has very strong antigenicity and specificity, and can be subjected to effective antigen-antibody reaction with an antibody generated in serum after sheep (Ovis aries) is infected with sarcosporidium tenella, so that diagnosis is realized; low-titer antibodies can be detected two weeks after infection, and the detection window period is obviously advanced; by adopting the ELISA detection method, a visual result can be directly obtained. The method breaks through the limitation of an infection window period, has the advantages of low cost, high sensitivity, simplicity and convenience in operation, no need of expensive instruments and analysis and the like, is suitable for field detection in primary laboratories and pastures, and is beneficial to industrial production and market popularization and application.

The invention provides an intelligent sensing data acquisition device and a data acquisition system, and relates to the technical field of data acquisition, and the intelligent sensing data acquisition device comprises an installation seat which is installed on a to-be-installed plane; the acquisition box is detachably mounted on the mounting seat, a plurality of mounting windows are formed in the acquisition box, and the mounting windows are used for mounting sensors; the sensor base is arranged in the acquisition box, the sensor base and the mounting window are correspondingly arranged, and the sensor base is used for being electrically connected with a sensor; the fastening piece is detachably arranged on the collection box, and the fastening piece is used for positioning and fixing the sensor. The sensor is flexibly disassembled and assembled by adopting a detachable design, the sensor can be flexibly arranged according to on-site detection requirements, and meanwhile, the mounting firmness of the sensor is ensured, so that sensing data can be stably and accurately acquired, and the sensor has the characteristics of simple structure and convenience in operation.

The invention relates to the technical field of semiconductor front channel detection, discloses a dark field detection system and method, and particularly relates to a detection system and a detection method. A calibration image of a bearing object is collected through an image sensor, whether the position of the image sensor has deviation or not is determined based on the spatial change relation between the calibration image and a pre-collected standard image, and the position of the image sensor is adjusted to a target position. An image sensor arranged on the basis of the target position is used for collecting images of the multiple components in the detection system, and the spatial positions of the multiple components are adjusted through the collected images. Compared with the prior art, the embodiment of the invention can solve the technical problems that the camera needs to be reassembled for observation in the processes of stability test, light path debugging and the like of the detection system in the prior art, and the camera is disassembled and assembled for many times and is complicated to debug.

The invention discloses a high-speed camera for semiconductor graph-free dark field detection and a detection method, and relates to the technical field of semiconductor detection, and the method comprises the steps: integrating an adjustable laser light source array, a high-performance TDI camera and a pre-scanning sensor, carrying out the brightness information collection of a pre-scanning region, and generating a pre-scanning reflectivity graph; sequentially irradiating a pre-scanning area with different illumination light through an adjustable laser light source array, generating a plurality of scanning images, and extracting features to generate a multi-modal differential feature field; predicting a main defect type in the region through a multi-mode illumination and defect feature collaborative enhancement algorithm to generate a corresponding illumination mode instruction; calculating an optimal TDI parameter through a TDI parameter dynamic modulation algorithm to generate a TDI camera control instruction; and scanning the wafer according to the two instructions, collecting original image data, inputting the original image data into the depth defect classification model, and outputting a defect information graph and corresponding information. The invention aims to solve the problem that semiconductor detection precision and efficiency are difficult to balance.

The invention provides a detection system and a detection method. The detection system comprises a first bright field detection light path, a second bright field detection light path, a dark field detection light path, a first platform and a second platform, when the dark field detection light path is used for performing dark field detection on the ith to-be-detected object at the dark field position, the target light path is used for performing bright field detection on the (i + 1) th to-be-detected object at the target position; the target light path is a first bright field detection light path or a second bright field detection light path, and the target position is a first bright field position or a second bright field position. The bright field detection light path and the dark field detection light path are separated, and each light path performs detection independently, so that mutual influence between light sources is avoided, and the detection precision of bright field detection and dark field detection is guaranteed. Besides, when bright field detection is carried out on one object to be detected, dark field detection can be carried out on the other object to be detected, that is, parallel detection on the two objects to be detected can be realized, and the detection efficiency is greatly improved.

The invention discloses a surface topography measurement device and method based on polarization phase shift. The device comprises a laser module, a polarization beam splitting and interference module, a polarization beam combining and imaging module, a polarization image acquisition module and a control and reconstruction module. The laser module comprises a monochromatic tunable laser, an optical fiber collimating amplifier and a polarizer; the polarization beam splitting and interference module comprises a polarization beam splitting prism, a first quarter-wave plate, a plane mirror and a second quarter-wave plate; the polarization beam combining and imaging module comprises a polarization splitting prism, a third quarter-wave plate and a telecentric lens; the polarization image acquisition module comprises a micro-polarization array camera and an image processing unit, and the micro-polarization array camera synchronously acquires four polarization interference images through single exposure; and the control and reconstruction module obtains measurement data of the surface topography of the to-be-measured object through a phase unwrapping algorithm and three-dimensional topography reconstruction. According to the invention, high-precision, large-field-of-view, portable and high-efficiency field detection can be realized.

The invention relates to the technical field of soil and underground water collaborative monitoring, and discloses a soil and underground water collaborative monitoring stratified sampling and field detection integrated system, which comprises a stratified sampling module, a field detection module, a collaborative data processing module and a central control module, the system is also integrated with an early warning feedback unit. Synergistic stratified sampling of soil and underground water is achieved through the stratified sampling module, samples at different depth layers are automatically obtained through a programmable depth control device and a multifunctional sampler, and meanwhile in-situ rapid detection is conducted by integrating a multi-parameter sensor and a spectral analysis unit through the field detection module. Key indexes are measured in real time, instant feedback and calibration of data are supported, and automation and data accuracy of the monitoring process are guaranteed; soil and underground water monitoring data are fused through the collaborative data processing module, and the risk early warning and decision support capability of the monitoring data is ensured.

The invention provides a tea tender shoot recognition method based on improved YOLOv11n in a complex environment, and the method comprises the steps: obtaining a field picture of a tea tender shoot, and making the obtained picture into a data set through image processing; constructing a YOLOv11n-based tea tender shoot identification network model in a complex environment, and improving the YOLOv11n model; testing a test set in a tender tea data set for the trained improved YOLOv11n model; and performing field detection on the tea tender shoots by using the trained network model. According to the method, the YOLOv11n model is improved, a CGConv feature extraction module, an improved head structure and a replacement bounding box loss function are introduced, local features, surrounding context features and global features can be fused, extraction and utilization of multi-scale information are achieved, and the method is suitable for large-scale application. The loss module can further improve the detection capability of the model on the small target by dynamically adjusting the weight, supporting more improved IoU loss, introducing a special strategy for small target detection and other improvements.