55results about How to "Improve detection results" patented technology

The invention discloses a remote sensing image change detection method based on fine tuning CLIP, and the method comprises the steps: obtaining an original dual-time remote sensing image, and carrying out the preprocessing of the remote sensing image; constructing an initial remote sensing image change detection model, and training the initial remote sensing image change detection model by using the training set to obtain a target remote sensing image change detection model; and performing change detection on the remote sensing image by using the target remote sensing image change detection model. According to the method, a large model CLIP pre-trained in a large-scale data set is introduced into a change detection task, and the method aims at guiding the model to pay attention to universal features of changing ground features, so that the generalization ability of the model is improved; according to the method, the CLIP is finely adjusted by adopting a visual prompt method, so that the CLIP can be well migrated in a remote sensing downstream task; in addition, when a loss function is constructed, Dice coefficient loss is introduced to enhance the sensitivity of the model to a positive example.

The invention relates to a pixel-level mask assisted camouflage target detection method based on edge prior guidance, and belongs to the technical field of computer vision. Firstly, Resnet is used as a backbone network to extract five layers of feature maps with different sizes, and the second layer of feature maps and the fifth layer of feature maps are used as input of an edge extraction module to obtain target edge priori with global semantic information. And then, target edge priori guides to extract target features of different scales from the second layer to the fifth layer of feature maps. Target features of different scales are fused in sequence from a high level to a low level through three context information aggregation modules, and a pixel-level mask is obtained through a Sigmoid function of the fused feature map with the same scale as the second-layer feature map. And finally, taking the feature map output by the second layer feature extraction module and the pixel-level mask with the same level scale as the input of a detection head, obtaining the category of the camouflage target and the coordinate of a detection frame, and assisting the detection of the camouflage target to obtain a finer detection result.

The invention relates to the technical field of valve detection, and discloses a multifunctional valve testing device and a use method.The multifunctional valve testing device comprises a water tank and a base installed in the middle of the water tank through two platforms, the two platforms are located in the middle of the front side and the middle of the rear side of the base respectively, and the base is provided with annular rails; the annular track is provided with a plurality of sets of fixing bases distributed at equal intervals, the annular track drives the multiple fixing bases to conduct fixed-distance conveying, the fixing bases circularly move around the end face of the base, and the positions of the fixing bases are switched in a reciprocating mode. The annular track can drive the fixed base to conduct fixed-distance conveying, so that the position of the workbench and the position of the multifunctional valve on the workbench can be automatically switched according to a set program. In the testing process, the valve can sequentially pass through a plurality of stations such as air inflation, water entry vibration, bubble detection and air pressure re-detection, the valve does not need to be manually and frequently carried, the testing efficiency is greatly improved, meanwhile, the detection time is prolonged, and the detection result is improved.

The invention discloses an improved YOLOv8-seg-based chip surface defect segmentation model and a training method and application thereof, a network of an improved YOLOv8-seg segmentation detection model is built, a Backbone module comprises a multi-layer ShuffleNetV2 network structure, a picture is subjected to multiple convolution processing, and three effective feature layers are extracted; the Neck module adopts a ConcatSDI network structure and an EMA attention mechanism network structure, the ConcatSDI network structure is used for carrying out multi-level feature fusion on an effective feature layer output by the Backbone module to obtain a feature map, and the EMA attention mechanism network structure is used for carrying out feature weighting on the feature map; the head module is a segmentation head and outputs final feature information based on an output result of the Neck module; and after the network structure of the improved YOLOv8-seg segmentation detection model is constructed, training, verification and testing are carried out. According to the method, the improved yov8-seg segmentation detection model is used for accurately segmenting and detecting the surface defects of the chip, and particularly, the detection precision can still be ensured when the method faces complex backgrounds and low-contrast defects.

The present invention relates to the field of building detection equipment, specifically to a precast concrete cover plate anti-crack bearing capacity detection device, comprising a support platform, and also comprising: a detection power box, the detection power box is connected to the support platform through a cylindrical rod; a thrust assembly, the thrust assembly is located in the detection power box; and a detection mechanism, the detection mechanism is respectively connected to the support platform and the cylindrical rod, and is installed in coordination with the thrust assembly; wherein the detection mechanism comprises an elastic pad, an observation tube, a protruding plate, an observation assembly and a pressurizing assembly. The precast concrete cover plate anti-crack bearing capacity detection device of the present invention has a novel structure, and can enable staff to understand the situation in a timely manner when imperceptible cracks appear during the detection process, thereby improving the detection results, ensuring the accuracy of the detection results, and providing convenience for the staff.

The present invention discloses a concrete steel bar bond strength detection device, which relates to the technical field of bond strength detection. The present invention comprises a bottom plate, on which four columns are fixed at equal intervals; a top plate, fixed on the upper end surfaces of the four columns, and the four columns are located near the four top corners of the top plate; a downward pressure cylinder, fixed in the middle position of the top plate; a lower pressure plate, fixed on the upper end surface of the bottom plate; four positioning assemblies, all fixed on the upper end surface of the bottom plate, and respectively located in the middle positions of the four sides of the lower pressure plate; a downward pressure assembly, slidably arranged on the four columns, and connected to the output end of the downward pressure cylinder; wherein, embedded concrete is placed on the lower pressure plate, and a plurality of steel bars are embedded in the embedded concrete; in the present invention, different steel bars are embedded in one embedded concrete, so that there is only one set of variables during detection, thereby comparing the concrete steel bar bond strength under different conditions.

The invention provides a colloidal gold test strip for detecting renal injury molecule 1 (kim-1) and a preparation method thereof, and aims to provide a bedside rapid diagnosis tool which is high in accuracy, strong in specificity, simple to operate and capable of rapidly reporting results. The test strip is formed by sequentially connecting a plastic bottom lining, a nitrocellulose membrane, a gold label pad, a sample pad and a water absorption pad, wherein a detection line and a quality control line are arranged on the nitrocellulose membrane. The detection line contains a KIM-1 monoclonal antibody, the quality control line contains a mouse anti-human IgG antibody, and the colloidal gold pad is coated with a colloidal gold labeled KIM-1 monoclonal antibody, so that a double-antibody sandwich structure is formed to specifically recognize human KIM-1. The preparation method comprises the following steps: spraying an antibody on the nitrocellulose membrane to form the detection line and the quality control line, preparing a colloidal gold antibody compound solution, and assembling each component. By optimizing the antibody concentration and a colloidal gold preparation process, the sensitivity and the stability of the test strip are ensured. The test strip is suitable for rapidly detecting kidney injury, and has important value for early diagnosis and treatment monitoring of acute and chronic kidney injury.

The present invention is applicable to the technical field of dust explosion-proof detection, and provides a dust explosion-proof performance testing device based on a limited space, including a pipeline component, wherein a detection component is snap-fitted on the outside of the pipeline component; the detection component includes a sealing member, two reciprocating members symmetrically slidingly fitted inside the sealing member, and a knocking member symmetrically slidingly fitted inside the sealing member, wherein the knocking member and the reciprocating member are slidingly fitted. The device solves the problem that, when the existing device is tested, some dust is easily adhered to the inner wall of the pipeline component for air circulation, and the outer wall of the pipeline component lacks a certain knocking structure, resulting in that when the dust concentration performance is detected inside the limited space, the air dust adhered to the inner wall of the pipeline component affects the later dust concentration performance detection data, thereby reducing the detection result of the dust explosion-proof performance test in the limited space and causing the dust explosion phenomenon to occur easily.

The invention provides a method for identifying fires in electric equipment based on a multi-parameter recursive graph, and relates to the technical field of electrical fire detection. The method comprises: using a smoothing filtering technology to pre-process monitoring video images, and creating a fire identification algorithm model based on artificial intelligence to detect and identify fire images in suspicious areas; the fire identification algorithm model identifies the hidden danger electrical monitoring data of the detection position, calculates the target feature vector corresponding to the trace features of the target video image, and builds a deep neural network sub-model and a power heterogeneous multi-parameter fusion model; then using a recursive algorithm module to arrange the power time series parameters into a recursive matrix, and constructing a graph convolutional neural network based on a multi-parameter recursive graph to extract features; using a fire warning judgment module to segment the suspicious unknown areas of the image sequence where the fire occurs, and timely predict and process them, and after identification and confirmation, automatically start a multi-parameter warning judgment program, and accelerate the issuance of warning signals by a background warning system.

This invention discloses a pedestrian detection method and system based on head center point-assisted feature comparison. The method includes: constructing a new CSP network architecture by adding a head branch, a feature comparison branch, a detection confidence enhancer, and a head-aware NMS module to the original CSP network architecture; constructing a training sample set, where the head center point is used as the positive sample for the head branch, and non-head center points within the head bounding box are used as the negative sample for the head branch; using the training sample set as input to the CSP network for training to obtain a pedestrian detection network model; supervising network learning during the training phase using head branch loss and feature comparison loss; and enhancing the detection results during the training and testing phases using the detection confidence enhancer and the head-aware NMS method. This invention improves the detection accuracy of CSP and reduces the probability of false positives and false negatives by introducing head center point, feature comparison, detection confidence enhancer, and head-aware NMS methods.

The utility model provides a cell viability measuring device in a cell culture dish, which belongs to the technical field of measuring devices and comprises a base, the base is connected with a detection sensor, a humidifier and a display instrument, one end of the detection sensor is provided with a detection head, and the base is connected with a control button, a humidity sensor, a sealing door and a temperature sensor. An electric turntable and a placement table are arranged in the base, an electric heating plate and a culture dish are arranged in the placement table, the placement table is connected with a partition plate, and the culture dish is connected with a connecting ring. The culture dish has the beneficial effects that the culture dish is prevented from being accidentally splashed into other culture dishes to cause pollution through the partition plate; meanwhile, fan blades are driven to rotate in the fixed cylinder while the partition plate rotates, so that air flow is better increased, the diffusion speed of temperature or humidity can be increased during heating or humidifying, efficiency is improved, diffusion uniformity is guaranteed, and detection accuracy is improved.

The invention relates to a visual closed-loop detection method, system and device for an automatic filling and sealing production line in the technical field of flexible filling and sealing automatic production lines, a hyperspectral camera and a 3D visual camera are adopted for data acquisition of a calibration information base, respective mathematical models are established, a coupling model is constructed, and finally whether the product is a qualified product or not is judged. Through multi-dimensional data fusion of the hyperspectral camera and the 3D camera, synchronous detection of spectral features and three-dimensional structure parameters of the potting layer is realized, single-dimensional judgment of traditional manual visual inspection is upgraded to multi-modal composite analysis, the detection precision and reliability are greatly improved, and particularly, whether a potting adhesive of a potting product is cured or not can be accurately judged.

The present invention relates to the field of data processing technology, and in particular to a data processing method and system for an integrated electric control device. The method comprises the steps of: obtaining the initial interval number of a histogram according to the number and discreteness of current data to obtain a histogram; obtaining the distribution uniformity of each interval of the histogram according to the abnormal possibility of each data in each interval of the histogram; obtaining the necessity of splitting each interval of the histogram according to the distribution uniformity, splitting each interval to obtain an updated histogram; obtaining the merging property of each interval of the updated histogram according to the similarity between the intervals of the updated histogram and the data difference between the interval and its similar interval, merging the intervals to obtain a final histogram; obtaining the abnormal score of each current data according to the final histogram, and performing abnormal detection. The present invention makes the HBOS algorithm more accurate in monitoring abnormal data and improves the results of abnormal detection.

This invention discloses a method, apparatus, device, medium, and program product for detecting contraband. The method includes: extracting features from X-ray security images based on a multi-branch, cross-stage local network to obtain multi-scale target feature maps. The multi-branch, cross-stage local network includes a shallow feature extraction module, multiple deep feature extraction modules, and a semantic feature extraction module. Each deep feature extraction module includes a first module and a second module. The first module uses a multi-branch structure to perform downsampling and extract features from receptive fields at different scales, and fuses the feature maps under each scale receptive field to obtain a multi-scale fused feature map. The second module uses a dual-path structure to perform cross-stage deep feature extraction and fusion processing on the multi-scale fused feature map. The multi-scale target feature map is input into a target detection network for detection to obtain the contraband detection results in the X-ray security image, thereby improving the efficiency and accuracy of contraband detection.

The invention discloses a table region recognition and extraction method and system for a full-frame table, and the method comprises the steps: generating a table document image data set, constructing a deep learning model which comprises an image sharing coding layer, a column region recognition decoding layer and a table total region recognition decoding layer, a picture passes through an image sharing coding layer and then is respectively input into two branches of a column area identification decoding layer and a table area identification decoding layer to obtain masks of a table and a column area; determining the ordinate of each row separation line through the pixel level accumulated value of the table document image; and carrying out model training, inputting the table image into the trained model, carrying out identification extraction on the text content of the table area in the document image by using optical character identification, and displaying the text content in an Excel format. According to the method, the problems that the text content in the table area in the image and other non-structural documents is difficult to recognize and extract and the recognition and extraction effect of the text content contained in the table is poor due to low table structure recognition accuracy are solved.

The invention relates to the technical field of data processing, in particular to a data processing method and system for an integrated electric control device. The method comprises the following steps: acquiring an initial interval number of a histogram according to the number and dispersion degree of current data to obtain the histogram; according to the abnormal possibility of each piece of data in each interval of the histogram, obtaining the distribution unicity of each interval of the histogram; according to the distribution unicity, obtaining the splitting necessity of each interval of the histogram, and splitting each interval to obtain an updated histogram; according to the similarity between the intervals of the updated histogram and the data difference between the intervals and the similar intervals of the intervals, the merging performance of each interval of the updated histogram is obtained, the intervals are merged, and a final histogram is obtained; according to the final histogram, obtaining the abnormal score of each current data, and carrying out the abnormal detection, thereby enabling the monitoring of the abnormal data through the HBOS algorithm to be more accurate, and improving the result of the abnormal detection.

The present invention belongs to the technical field of moisture detection devices, and specifically relates to a moisture detection device based on hyperspectral remote sensing, including a detection platform, an auxiliary flattening mechanism and an auxiliary drying mechanism are arranged on the upper side of the detection platform, a bracket is fixedly connected to the outer wall of the detection platform, and a detection head is detachably connected to the bracket, and the detection head is located directly above the detection platform. The moisture detection device based on hyperspectral remote sensing uses the auxiliary flattening mechanism to drive the output shaft of the motor to rotate the rotating seat, and the rotating wheel moves in the annular groove. In the process of passing through the hemispherical block, the rotating seat will move up and down, thereby causing the detection object in the detection placement box to shake, assisting the detection object in the detection placement box to be placed flat, and improving subsequent detection results. At the same time, after the detection is completed, the detection placement box can be in a tilted state, which is convenient for discharging the detection object in the detection placement box, and can dry the residual moisture in the detection placement box.

The utility model discloses a latex glove air tightness detection device, which relates to the technical field of latex glove production, and comprises a detection table, the left side of the top of the detection table is fixedly connected with an air pump, the output end of the air pump is fixedly connected with an inflation hose, and the right end of the inflation hose is provided with a sealing installation assembly. A connecting pipe is inserted into the right end of the sealing mounting assembly, a transparent water tank is fixedly connected to the right side of the top of the detection table, the bottom of the outer wall of the connecting pipe penetrates into the transparent water tank, a water inlet assembly is arranged on the front side of the top of the transparent water tank, and a supporting table is fixedly connected to the bottom of the sealing mounting assembly; a control switch is arranged on the left side of the front face of the detection table. Water is injected into the transparent water tank, then the latex gloves are installed in the sealing installation assembly, the air pump is started to blow air to the latex gloves in the sealing installation assembly, and then whether bubbles are generated in the transparent water tank or not is observed.

The invention relates to the technical field of railway vehicle maintenance, in particular to a vehicle wheel set detection device which comprises a supporting seat, a lifting assembly is fixedly mounted on the supporting seat, a supporting sleeve is fixedly mounted at the upper end of the lifting assembly, a through hole is formed in the supporting sleeve, a driving motor is fixedly mounted on the lower surface of the supporting sleeve, and the driving motor is connected with the lifting assembly. The output end of the driving motor is fixedly provided with a driving wheel through a coupler, the supporting seat is fixedly provided with a positioning sleeve, the supporting seat is fixedly provided with a protective cover, and the protective cover is fixedly connected with a fixing block. The device integrates laser array detection and a double-CCD visual verification technology, double verification of laser quantitative detection and visual qualitative verification is achieved, and the detection accuracy is improved. Single detection misjudgment is avoided, and the detection precision and the result credibility are improved; the whole detection process is automatic, manual intervention and operation errors are reduced, and labor intensity is reduced; the device is scientific in structural design, adapts to wheel sets of multiple vehicle types, guarantees the consistency of detection standards, and gives consideration to the use safety and environmental interference resistance.

The utility model provides a stator inner diameter detection device, which relates to the technical field of stator detection tools, and comprises a cylindrical axial rod and a detection assembly fixed at one end of the axial rod, the detection assembly comprises a fixed rod and threaded rods respectively arranged at two ends of the fixed rod, and the axial rod is vertically fixed in the middle of the fixed rod. The ends of the two threaded rods are in threaded connection with abutting pieces respectively. The device has the advantages that the inner diameter of the stator can be quickly detected, and the assembly clearance of the screw pump can be conveniently controlled.

The invention discloses a biomarker detection kit and a detection method for prognosis evaluation of multiple myeloma, and relates to the technical field of biological medicine detection.According to the kit, circMYC and circSETD2 bicyclic RNA combination in serum exosomes serves as a core prognosis marker, and the kit contains an exosome solid phase capture reagent, sulfydryl modified silicon dioxide magnetic beads and the like; the isothermal amplification detection reagent contains a specific SDA primer pair and the like; a double quality control reagent containing a CD63 peptide fragment marker and the like; positive control and negative control are provided; the serum exosome bicyclic RNA is selected as a core prognostic marker, a complete detection system is constructed, an exclusive reagent efficiently enriches and purifies the exosome, the detection specificity and accuracy are improved, a reliable basis is provided for prognostic layering, integrated operation is achieved, the process is simplified, clinical application is achieved, quantitative analysis is matched with a clear interpretation standard, and the detection efficiency is improved. The reagent is stable and convenient, the result is comparable, and the clinical application value is improved.

The invention relates to the technical field of biomedical examination, and particularly discloses a proprotein convertase subtilisin / kexin type 9 standard product protective agent, a detection kit for determining proprotein convertase subtilisin / kexin type 9 and application of the detection kit. The protective agent is prepared from the following components in percentage by weight: 1 percent to 3 percent of protein, 0.5 percent to 2 percent of amino acid, 1 percent to 5 percent of a surfactant, 0.01 percent to 0.5 percent of an antioxidant, 0.01 percent to 0.2 percent of a de-foaming agent, 0.01 percent to 0.2 percent of EDTA (Ethylene Diamine Tetraacetic Acid), 0.05 percent to 0.2 percent of a preservative and the balance of a buffer solution. The proprotein convertase subtilisin / kexin type 9 standard product protective agent provided by the invention can significantly improve the stability of the proprotein convertase subtilisin / kexin type 9, prolong the effective period of the detection reagent, and ensure the accuracy and repeatability of the detection result. The detection kit provided by the invention is short in detection time, good in precision, high in sensitivity and good in reagent uniformity.

The invention belongs to the technical field of moisture detection devices, and particularly relates to a moisture detection device based on hyperspectral remote sensing, the moisture detection device comprises a detection table, an auxiliary flattening mechanism and an auxiliary drying mechanism are arranged on the upper side of the detection table, a support is fixedly connected to the outer wall of the detection table, and a detection shooting head is detachably connected to the support; and the detection shooting head is positioned right above the detection table. According to the moisture detection device based on hyperspectral remote sensing, an auxiliary flattening mechanism is used, so that an output shaft of a motor drives a rotating seat to rotate, a rotating wheel moves in an annular groove, and the rotating seat moves up and down in the process of passing through a hemispherical block, so that a detected object in a detection placement box is shaken; the detection object in the auxiliary detection placing box is flatly placed, the subsequent detection result is improved, meanwhile, after detection is completed, the detection placing box can be in an inclined state, the detection object in the detection placing box can be conveniently discharged, and residual water in the detection placing box can be dried.

The utility model relates to a detection mechanism for a series welding machine, which comprises a tool frame body, at least two groups of detection assemblies and a display element, the two groups of detection assemblies are oppositely arranged on the upper half part and the lower half part of the tool frame body, and the display element is fixedly connected to the tool frame body through a second support. The display element is electrically connected with the two groups of detection assemblies, the detection assemblies are used for detecting the state of the battery string conveyed on the conveying assembly, and the display element is used for displaying the detection result, so that automatic detection of the battery string is realized, the detection efficiency is effectively improved, the detection quality is improved, and the situation of missing detection or false detection is effectively avoided. The tool frame body comprises a frame structure, the detection assembly and the display element are both fixedly connected to the frame structure, a supporting column is arranged at the end, close to the ground, of the frame structure, and pulleys are further arranged at the bottom, close to the supporting column, of the frame structure.

The present application relates to the technical field of plantago asiatica L. genetic engineering, and specifically provides a plantago asiatica L. real-time fluorescent quantitative PCR internal reference gene and application thereof. The tender leaves, old leaves, stems, flowers and roots of plantago asiatica L. are selected as experimental materials, candidate internal reference genes are screened through plantago asiatica L. transcriptome, then the gene expression stability is analyzed through real-time fluorescent quantitative PCR analysis, melting curve analysis, amplification efficiency analysis and various software (geNorm, NormFinder, BestKeeper, Delta Ct and RefFinder) analysis, and the internal reference genes suitable for gene quantification of different tissue parts of plantago asiatica L., i.e. eIF and Cycl genes, are screened, and the sequences are shown as SEQ ID NO. 1 and SEQ ID NO. 2. The internal reference genes with stable expression are obtained, are suitable for analyzing the expression amount of genes in different parts of plantago asiatica L., make up for the current situation that plantago asiatica L. lacks internal reference genes, and provide a basis for gene expression amount and function analysis in plantago asiatica L.

The invention relates to the technical field of photovoltaic power generation, in particular to a distributed intelligent photoelectric photo-thermal energy storage station system, and solves the problems that a test result is not accurate, materials are easy to waste, stability during testing cannot be kept and the like due to the fact that a single electrical performance test before or after assembly is usually carried out when a photovoltaic panel is subjected to a production test in the prior art. The distributed intelligent photoelectric photo-thermal energy storage station system comprises an electrical performance test device, the electrical performance test device comprises a loading positioning mechanism, a steering power connection mechanism and a distributed lighting mechanism, the inner side of the loading positioning mechanism is provided with the steering power connection mechanism, and one end of the steering power connection mechanism is provided with an automatic plugging mechanism. According to the invention, the electrical performance test is carried out before and after the assembly of the photovoltaic panel, the accuracy of the test result can be effectively improved, the material waste is reduced, the photovoltaic panel is vertically placed, the stability of the test can be kept, and the test accuracy is improved.

The application provides a reagent and kit for quantitatively detecting chicken egg protein allergen specific IgE. The kit comprises biotin-labeled chicken egg protein allergen and a preservative solution, the biotin-labeled chicken egg protein allergen comprises Sulfo-NHS-biotin-labeled ovomucin, Sulfo-NHS-LC-biotin-labeled ovalbumin, Sulfo-NHS-biotin-labeled ovotransferrin and Sulfo-NHS-PEG4-biotin-labeled lysozyme with a mass ratio of (0.5-1):1:(0.5-1):0.5, and the preservative solution is a phosphate buffer solution containing 1-2 wt% BSA, 1-3 wt% glycerol and 2-5 wt% mannitol. The reagent has higher detection rate, detection sensitivity and reliability, the detection result has higher consistency with clinical symptoms, and the batch consistency of the finished product is better.

The present invention discloses a textile fabric elasticity detection device that is convenient for comparison, comprising a processing base plate, a hydraulic push rod vertically mounted on the processing base plate, two groups of hydraulic push rods, a detection device disposed on the top of the hydraulic push rod, the detection device comprising a power unit, a winding adjustment unit, a transmission unit, a winding unit and an angle adjustment unit, the power unit comprising a servo motor, the servo motor being located on a motor frame, the motor frame being located on the processing base plate, the servo motor being connected to a power shaft, and a connecting rod being disposed at one end of the power shaft away from the servo motor. The textile fabric elasticity detection device that is convenient for comparison pushes the bottom of the limit plate to tilt by rotating the internal threaded rod of the angle adjustment unit, pushing the tensioning roller, thereby changing the position of the tensioning roller on the mounting column, and then because the horizontal position of the initial roller remains unchanged, the tightened fabric is no longer pulled horizontally, but is tightened obliquely, thereby improving the detection range of the fabric elasticity.

The utility model discloses a flow cytometer signal collection device. The flow cytometer signal collection device includes base, optical fiber subassembly and filter component, optical fiber subassembly is located on the base, including optical fiber structure, collimation structure and reflection structure, collimation structure is located on the base in the activity, and is located between optical fiber structure and reflection structure, collimation structure has the activity stroke of being close to optical fiber structure to be away from reflection structure and being away from optical fiber structure to be close to reflection structure, filter component is located on the base, and is located one side of optical fiber subassembly, and the filter channel is formed on filter component, and filter channel and reflection structure correspondingly are arranged, so that the fluorescence that reflects through reflection structure can be conducted into filter channel along the layout direction of filter channel. The utility model makes the light that filter component obtains more concentration, and electrical component can better carry out the detection, and effectively promotes the detection result, and the applicability of flow cytometer signal collection device for different wavelength fluorescence is better also.

The utility model belongs to the field of photoelectric detection, and provides a liquid concentration photoelectric online detection sensor which comprises a detection shell I and a detection shell II, communicated mounting cavities are arranged in the detection shell I and the detection shell II, a liquid inlet is arranged on a top plate of the detection shell I, a liquid stagnation pipe is arranged in the liquid inlet, and a liquid outlet is arranged in the liquid stagnation pipe. The liquid stagnation pipe is a hollow pipe body; the detection prism is located in the mounting cavity, the top of the detection prism is attached to the bottom of the liquid stagnation pipe, a measuring light source externally connected with a power source is arranged in the mounting cavity, a light source opening of the measuring light source faces the liquid stagnation pipe, a photoelectric receiver is arranged right opposite to the measuring light source, and the bottom of the detection prism is fixed through a supporting assembly; according to the utility model, the detection prism is additionally arranged in the sensor, so that the accuracy of a concentration detection result of colored and turbid liquid is improved.