49results about How to "Avoid visual errors" patented technology

The invention relates to a foundation bearing capacity filling packing in-situ detector, belonging to the geotechnical engineering mechanics field. which comprises: a casing, a crossbeam, a load sensor, a transmission shaft, a sounding rod, a sounding head, a data acquisition processor, a control display panel, a communication interface and a power supply box; The invention is characterized in that: the transmission shaft, the sounding rod and the sounding head is connected with the bottom of the load sensor fixed in the inner cavity of the casing; when the sounding head is inserted into soilmass, the load sensor generates a changeable signal, the signal is sampled, amplified, calculated and displayed by the data acquisition processor, and sent to the upper machine through the communication interface for intelligentizing processing. The invention is distinguished from the penetration test apparatus, sand cone method and nucleon density apparatus used in current engineerings, can solves the following problems in existing technologies: 1, in-situ disturbing dehydration caused by drilling soil sampling, 2, radiation pollution of the nucleon density apparatus, 3, the time and labor consumed in digging pit and filling sand, 4, large error in spring force measuring element, 5, visual error in rule read value and impossibility for keeping data.

The invention discloses an image description method for fusion of visual context, comprising the following steps: 1) preprocessing; 2) image description label preprocessing; 3) feature extraction; 4) mean value pooling; 5) convolution and mean value sampling Pooling; 6) Obtaining detected image entities; 7) Obtaining entity attributes; 8) Convolution; 9) Obtaining entity attribute features; 10) Convolution; 11) Convolution; 12) Convolution; 13) Acquiring entities and attributes Relationship; 14) The relationship between entities and attributes; 15) LSTM training; 16) Solve the exposure bias; 17) Reduce the dimension; 18) Normalize; 19) Get the description sentence of the current image, that is, the model; Descriptive statement; 21) Test and verify the training effect and performance of the model. This method can ensure the accuracy of image feature extraction, avoid visual errors, make the generated description more fluent to conform to human grammar rules, and obtain higher scores for evaluation indicators.

The invention discloses a flow controller for intravenous infusion. The flow controller for intravenous infusion comprises a pliers body. The pliers body comprises two half pieces which are hinged together. The portions, on the rear side of the hinge point, of the two half pieces are used as hand-operated clamping handles. The portions, on the front side of the hinge point, of the two half pieces form a right clamp body and a left clamp body. The middle portion of the left clamp body and the middle portion of the right clamp body are symmetrically provided with arc-shaped grooves which define an infusion set hose clamping cavity. The position, on one side of the infusion set hose clamping cavity, of the left clamp body or the right clamp body is provided with an infrared sensing emitter, and the position, on the other side of the infusion set hose clamping cavity, of the right clamp body or the left clamp body is provided with an infrared sensing receiver. An electric control unit and a setting and displaying module are arranged on the left clamp body or the right clamp body. The electric control unit is connected with the setting and displaying module, the infrared sensing emitter and the infrared sensing receiver respectively. The flow controller for intravenous infusion can accurately display the flow rate in the infusion set hose, and has the two functions of manually and automatically adjusting the flow rate in the infusion set hose; no mechanical adjustment in the manual and automatic switching process is needed, so the flow controller for intravenous infusion is very convenient to use.

The invention relates to a warp knitting fabric defect detection method based on wavelet contourlet transformation and visual saliency. The method comprises the steps: selecting a fundamental wave, and constructing a wavelet transformation filter; performing wavelet decomposition for a warp knitting fabric image to be detected, and obtaining approximate feature sub-graphs and detail feature sub-graphs; performing Gauss difference among the approximate feature sub-graphs and the detail feature sub-graphs so as to obtain approximate feature difference sub-graphs and detail feature difference sub-graphs; performing normalization processing for the feature difference sub-graphs, and performing addition and mean value treatment to obtain approximate feature saliency graphs and detail feature saliency graphs; utilizing a non-subsample direction filter bank to perform convolution filtering for the detail feature saliency graphs to obtain detail feature direction sub-band coefficients, and selecting a higher sub-band coefficient with higher energy according to the energy theory to reconstruct the detail feature saliency graphs; and performing segmentation for the approximate feature saliency graphs and the reconstructed detail feature saliency graphs, adding each segmented images after processing the segmented images, and then performing defect determination. The warp knitting fabric defect detection method based on wavelet contourlet transformation and visual saliency can improve the defect detection accuracy.

The invention relates to a method for eliminating a visual error between a forward-looking infrared device and a head-up display. By calculating a deviation of a visual axis of the forward-looking infrared device relative to a visual axis of the head-up display according to attitude data of a carrier aircraft, performing translation and shear processing on an infrared image output by the forward-looking infrared device and sending a processed infrared image to the head-up display to display, the visual error caused by mounting positions of the forward-looking infrared device and the head-up display can be solved, and an image displayed by the forward-looking infrared device on the head-up display is guaranteed to be consistent with an outside scene image seen by a driver via the head-up display.

The invention discloses a visual error elimination method of a two-dimensional vector solid line in a three-dimensional scene, which mainly solves the intermittent disappearance phenomenon caused by the thin vector solid line existing in the process of fitting and rendering the two-dimensional linear symbol to the three-dimensional terrain surface, vector The aliasing and aliasing phenomenon of the solid line boundary, the phenomenon that the width of the vector solid line does not change with the zoom of the scene, etc. It includes the following main steps: according to the width of the vector solid line, it is planarized; using the projection inversion method to calculate the coverage of screen pixels in two-dimensional space; according to the spatial position relationship between the pixel's two-dimensional space coverage and the vector solid line area Classify the pixels, and perform multi-sampling processing on the part of the pixels that intersect with the area boundary of the vector solid line; judge whether the sampling point is inside the area of ​​the vector solid line, determine the coloring of the sampling point according to the judgment result, and use the sampling point The color blend value replaces the screen color buffer value for the corresponding pixel.