38 results about "Scale sliding" patented technology

The invention provides a method for detecting gestures. The method includes steps of detecting a predefined motion mode to determine a region-of-interest to be detected; detecting in the region-of-interest according to a scheme of a multi-scale sliding window, extracting local mean features from skin color membership images based on window images at first, classifying the features through a pre-trained Ada-Boost classifier, further extracting feature points on the basis of a gray-scale image of the window images processed via the Ada-Boost classifier, and then classifying the point-pair features through a random forest classifier; and clustering different types of target gesture windows and outputting accurate positions and shapes of the gestures. The invention further provides a system for detecting the gestures. The method and the system for detecting the gestures are simple, rapid and stable in implementing, and have the advantages of real-time performance, interference resistance, high identification precision and the like.

A system for automated geospatial image analysis comprising a deep learning model module and a convolutional neural network serving as an automated image analysis software module. The deep learning module receives a plurality of orthorectified geospatial images, pre-labeled to demarcate objects of interest, and optimized for the purpose of training the neural network of the image analysis software module. The module presents marked geospatial images and a second set of unmarked, optimized, training geospatial images to the convolutional neural network. This process may be repeated so that an image analysis software module can detect multiple object types or categories. The image analysis software module receives a plurality of orthorectified geospatial images from one or more geospatial image caches. Using multi-scale sliding window submodule, image analysis modules scan geospatial images, detect objects present and locate them on the geographical latitude-longitude system. The system reports the results in the requestor's preferred format.

The invention discloses a welding technics method of large size slipper mill barrel entrance and exit sliding ring, which adopts a strict web plate (2) and a sliding ring board shell ring (1) to manufacture welding technics, throttle, exit sliding ring welding quality and welding distortion to control, thus ensuring the geometry size accuracy and product quality of slipper mill barrel entrance and exit, and ensures the formation tolerance of the web plate (2) and the sliding ring board shell ring (1) and achieves the design requirement, which has strong operating performance and low production cost.

The present invention provides a road traffic sign detection and identification method. The method comprises the steps of: extracting a region of interest; performing multi-scale sliding traversal; merging image features; and identifying a traffic sign. The method specifically comprises: performing extraction of a region of interest of the traffic sign on an input to-be-detected image; convertingthe to-be-detected image into a gray-scale image; constructing a binary mask image of the region of interest of the traffic sign; performing multi-scale sliding traversal on the gray-scale image and the binary mask image to obtain position coordinates of a detected target; merging the extracted local texture features, local image region features and global features of the to-be-detected image; andclassifying and identifying the merged image features by using a classifier. The present invention relates to an electronic device and a readable storage medium for performing the road traffic sign detection and identification method; and the present invention also relates to a road traffic sign detection and identification system. The technical scheme of the present invention has a high detection rate, a high identification rate, a fast calculation speed, a low false detection rate and good robustness.

The invention discloses an adjustable water-injection ablation electrode needle, and belongs to the field of operating instruments for electrosurgery and the field of operating instruments for minimally invasive ablation. The electrode needle comprises a needle rod, a handle, an adjusting knob, an adjusting sleeve, a scale sliding block, a needle rod fixed seat, a water pipe fixed seat, a water pipe and radio frequency cables, wherein the tip end of the needle rod is composed of a triangular blade which is easy for puncture, and a thin hole is formed in the needle rod; an annular groove is formed in the handle, and a length-adjusting graduation ruler is arranged on the edge of the annular groove; an outer screw thread is formed in the adjusting knob, and the adjusting knob is composed of a plurality of edges provided with rounded corners; the adjusting sleeve and the scale sliding block are bonded by virtue of glue; an inner screw thread, which is matched with the adjusting knob, is formed in the scale sliding block; a scale indicating line is printed on the scale sliding block; the needle rod fixed seat and the needle rod are connected by virtue of glue; the water pipe fixed seat and the water pipe are connected by virtue of glue; and the radio frequency cables are welded on the needle rod. The adjustable water-injection type ablation needle disclosed by the invention is convenient to operate and clear in scale, and the ablation needle is convenient for doctors to observe in an operation.

The invention discloses a method for dividing a level set image based on characteristics of a neighborhood probability density function, which mainly solves the problem existing in the prior method that the variability of texture elements of a natural image in texture scale and texture direction cannot be effectively processed. The method comprises the following steps of: initializing a level set function; setting an initial value and a maximum value for the iteration number of the level set function; working out a characteristic set of the neighborhood probability density function; working out characteristics of a subset probability density function; working out a similarity vector; according to an evolution equation of the level set, updating the level set function; judging whether to enter a cycle, if so, working out the characteristics of the subset probability density function, the similarity vector and the level set function again; otherwise stopping the cycle, and using a zero level set of the level set function as a segmentation border of an input image. Because the texture elements of the natural image are fit by a multi-scale sliding window, the variability of the texture elements in texture scale and texture direction can be better processed, so that the method can be applied to the segmentation of the natural texture image.

The invention discloses a fast pedestrian detection method with the combination of static bottom characteristics and motion information. For a single image frame in a video sequence to be detected, amulti-scale sliding window method is used to extract a detection window of a specified size, a fast pyramid characteristic calculation method is used to calculate the static layer characteristics in each detection window, the static bottom layer characteristics comprise a direction gradient, a gradient amplitude and a color channel, the pre-classification of a pedestrians and no pedestrian is carried out on each detection window based on the static bottom characteristics of an image in each detection window. The specific motion speed characteristics of the pedestrians are used, through a detection window pixel-difference-averaging method, a detection window which is mistakenly detected to be a pedestrian in the pre-classification is removed, and finally the position of an image in a detection which is detected to be a pedestrian is the position of the pedestrian in the image frame. According to the method, the accuracy of detection is improved while the false detection rate is reduced.

The invention provides a target tracking method based on the application of an unmanned boat. The method comprises the following steps of acquiring an image sequence of a target, and searching a maximum response position by utilizing a pre-set filter f(z)=WTZ within a preset range with the position of the target at the previous frame as a search center, wherein the maximum response position is adopted as the candidate position center of the target at the current frame; sampling through a multi-scale sliding window at the periphery of the candidate position center to obtain a plurality of candidate frames, calculating the scores of all candidate frames by using a structured SVM classifier, and adopting a candidate frame largest in score to be the prediction result of the target at the current frame; judging whether the target is shielded or not; if the target is shielded, not performing the online learning and not updating the parameters of the pre-set filter and the parameters of the structured SVM classifier; if the target is not shielded, updating the parameters of the pre-set filter and the parameters of the structured SVM classifier; repeating the step S1 to S5 until reaching the last frame of the image sequence. The method can realize the long-time and stable tracking function.

The invention discloses an image retrieval method based on depth learning and approximate target location. The method comprises the steps of focusing on searching the approximate object region in theimage; extracting the corresponding local CNN features by using multi-scale sliding windows on the image. With these small feature blocks, the feature vectors of the image can be obtained, which can meet the requirements of image retrieval. After processing the features, optimized image representations with several weights have been prepared for image retrieval and rearrangement; a new frame called approximate object position is constructed to search the most similar domain which can be retrieved from the queried image. For the retrieval of a specific object, a series of small feature blocks are used to locate the approximate object on the database image, and the retrieval process is completed by the region located. The invention fully utilizes the powerful ability of depth learning, optimizes and innovates based on CNN, improves image retrieval accuracy and reduces redundancy.

The invention discloses a moving ground flatness detecting device, a detecting system and a detecting method. The moving ground flatness detecting device comprises a main shaft which is vertically arranged on the ground to be detected, a tripod, a rotary drive and a laser range finder. A cross bar of the tripod is vertically fixed on the main shaft and can horizontally rotate around the main shaft. One end of the cross bar is a scale sliding rail. The rotary drive is fixed on the main shaft and is connected the cross bar to drive the cross bar to horizontally rotate. The laser range finder is arranged on the scale sliding rail of the cross bar, and can horizontally slide along the scale sliding rail to measure the height of the surface at different positions of the ground to be detected. According to the moving ground flatness detecting device provided by the invention, the rotary drive drives the scale sliding rail and the laser range finder to rotate; the height of the surface at different positions of the ground to be detected is measured at different measurement radius; the relative height difference among all points in the measured plane range can be determined; and the flatness of the ground to be detected is effectively determined.

The invention discloses a direction finding and target recognition method based on a passive phased array antenna. The method is based on an annular passive phased array antenna structure, and a digital beam forming method is adopted, so that the radiation source detection distance and the angle measurement precision are improved; and aiming at a multi-period radar signal received in an actual passive radar processing system, based on an idea of short-time Fourier transform, period characteristic information of the signal is extracted by adopting a variable-scale sliding window short-time Fourier transform FrFT method, and finally a radiation source is identified, so that the problems that a traditional method is large in data amount and high in operation load, and time domain period information of the signal is easily submerged in spectral line distribution of a transform domain are solved.

The invention discloses an ultraviolet spectrum water quality abnormality detecting algorithm based on multi-scale sliding window principal component analysis. The method comprises the following steps: 1) solving a wavelet transformation dimension L, a window length N, a Cusum control limit H1 of each scale and a wavelet reconstruction data Cusum control limit H' according to historical data; 2) acquiring on-line spectral data, and waiting for the window N filled with the data; 3) performing baseline correction and standardized preprocessing on the spectral data; 4) performing MSPCA on the spectral data, and selecting a principal component number according to a threshold value method; 5) performing each scale abnormality detection based on a Cusum control chart; 6) abnormal wavelet scale combination and reconstruction, and performing PCA calculation on the reconstruction data; 7) performing reconstruction data abnormality detection based on the Cusum control chart, and generating a water quality report. According to the method, the traditional principal component analysis ultraviolet water quality abnormality detection method is improved, so that the method can dynamically adapt tothe water quality background fluctuation, multi-scale water quality abnormality detection can be performed, and the detection accuracy of dynamic change water quality detection is increased.

The invention discloses a rail vehicle door sub-health state recognition method based on time series data mining. A multi-scale sliding window discrete character algorithm is used to carry out data mining on rotation speed, torque and current of a rail vehicle door motor. Through calculating a distance from a template curve in a normal state as a characteristic and using principal component analysis to reduce a dimension, redundant information is removed, and a low dimensional characteristic with good classification performance is obtained. Then, a hierarchical sub-health state recognition model is used to identify a variety of sub-health data layer by layer from coarse to fine, finally the identification of a sub-health state in the process of opening and closing a door is realized, the characteristic of vehicle door motor time series data can be effectively extracted, common sub health in the process of opening and closing the door is accurately identified, and the method has a good application prospect.

The invention discloses a wall plate part bending and warping deformation measurement device.The wall plate part bending and warping deformation measurement device comprises two main scales, a main scale sliding block is arranged between the two main scales and provided with a main scale measurement meter, the two main scales are vertically connected with an auxiliary scale and a sliding scale, the auxiliary scale is provided with an auxiliary scale measurement meter, the sliding scale is provided with sliding sleeves, and the sliding sleeves are provided with sliding scale metering meters.By means of the wall plate part bending and warping deformation measurement device, on-site online measurement of bending and warping deformation of a part profile can be achieved.A measurement method includes the specific steps that 1, part length, part width and the planned measurement direction are determined; 2, the measurement area is determined; 3, the measurement device is calibrated; 4, sliding supporting feet and the sliding sleeves are arranged on the measurement area, and sliding scale supporting feet are located on the measurement area; 5, readings of the measurement scales are recorded, and bending and warping deformation of the part profile is calculated.The wall plate part bending and warping deformation measurement method is simple in operation.

The invention discloses a multi-specialized target detection algorithm based on deep learning. A detection model framework of the target detection algorithm mainly comprises a RoI generation module and a multi-specialized channel module, wherein the target detection algorithm obtains an augmented RoI set through a multi-scale sliding window and selective search, and takes over the multi-scale sliding window in an exhaustion mode to generate a dense RoI set; a plurality of special channel networks are designed, each special channel network is responsible for detecting one type of RoI, and eachspecial network is provided with a full connection layer in the same form. According to the method, two strategies are used for improving the performance, the two strategies enhance the performance ofthe detection model in different aspects, and multiple special channel detection related to the shape category is designed; an augmented RoI dataset is further generated using a selective search andprovided to the network for training.

The invention discloses an egg-shaped pressure-resistant shell rapid positioning measurement apparatus. The apparatus comprises a base, the base is provided with a pair of clamping centering mechanisms, each clamping centering mechanism clamps one end of an egg-shaped pressure-resistant shell, a measurement mechanism is installed on the base, the measurement mechanism comprises a support plate apparatus arranged on the base, the upper end of the support plate apparatus is provided with a through hole, the through hole is internally provided with a rotating shaft, the rotating shaft is provided with a circular positioning plate, the circular positioning plate is provided with two circumferentially uniformly arranged limiting holes, a support plate is provided with four circumferentially uniformly arranged positioning pin mechanisms used in pair with the limiting holes, the tail end of the rotating shaft is connected with a limiting plate, the limiting plate is provided with a groove, and the groove is internally provided with a calibrated scale sliding along the groove. According to the invention, active adjustment mechanisms, a rotation scale mechanism and a wide type measurement head are employed, the egg-shaped pressure-resistant shell is conveniently and rapidly clamped and positioned, it is ensured that the axis of the egg-shaped pressure-resistant shell is horizontally arranged, the measurement precision of a short shaft and a long shaft of the egg-shaped pressure-resistant shell is enabled to be high, and the efficiency is high.

The invention discloses a motorcycle frame rigidity performance test device. The device comprises a pedestal. The pedestal is provided with a vertical pressure cylinder and a horizontal pressure cylinder respectively. Tail ends of the two pressure cylinders are connected with a vertical pressure sensor and a horizontal pressure sensor respectively. The other end of the vertical pressure sensor is connected with a rear pressure plate seat. The other end of the horizontal pressure sensor is connected with a horizontal slide carriage. A headstock rod support is fixed on the horizontal slide carriage. A headstock rod shaft, which is used to fix a front fork of a motorcycle frame, is installed on the headstock rod support. Side edges of the rear pressure plate seat and the horizontal slide carriage are provided with grating scale sliding plates. A left support seat and a right support seat, which are used to fix a rear fork, are fixedly arranged at a position which is close to a vertical rack on a horizontal rack of the pedestal. According to the invention, an actual road frame rigidity performance test before a new product is put in production can be abandoned. A structure is simple and reasonable. A demand for testing and evaluating frame intensity and rigidity performance during a development process of the motorcycle frame can be satisfied. By using the device of the invention, the new product can be rapidly made into markets.

The invention relates to a trademark image retrieval method based on multi-scale regional characteristic comparison The method comprises the following steps of (1) segmenting an image by using a multi-scale sliding window to obtain a series of windows and then extracting the characteristics of a window image by using a LBP operator, (2) carrying out global inter-scale characteristic window matching on the extracted characteristics to obtain a similar window pair Ai:Bj, (3) eliminating a mismatch in the similar window pair by using a spatial transformation model and screening scale-space consistent similar window, (4) segmenting the screened similar window pair by using an adaptive threshold matrix T to obtain a similar region ROI, (5) performing local window characteristic matching in thesimilar region ROI, and (6) outputting a retrieval result according to a match result of the local window characteristic in the similar region ROI. The ranking of a trademark image retrieval result obtained by the method is consistent with that of a human visual image judgment, the retrieval is fast, and the system robustness is good.

The invention discloses a human body detection method based on sample transfer learning. The human body detection method based on sample transfer learning comprises the following steps: multi-dimensional human body features of positive samples and negative samples in a training set are extracted; a support vector machine is used for training to obtain an initial human body detector; the human body detector is divided into a plurality of small grids, wherein each small grid represents a small component, and the weight of the corresponding grid is used for representing the small component; the weight of the grids is updated; human body detection is conducted on an image to be detected in a scanning mode through a multi-scale sliding frame, and therefore a human body detection result is obtained. According to the human body detection method based on sample transfer learning, the detector can be structurally adjusted in a self-adaptive mode so as to be matched with the structure of the samples in a detection frame and reckon in matching loss into a scoring function, and therefore the human body detection method can be used for conducting structural transferring on each sample in a self-adaptive mode and processing shape change in a human body and other conditions.

The invention provides a stratum mobility simulating device. The stratum mobility simulating device comprises a constant speed and constant pressure pump, an intermediate container, a high temperature and high pressure sample allocation device and a model mechanism, wherein one end of the intermediate container is connected with the constant speed and constant pressure pump, one end of the high temperature and high pressure sample allocation device is provided with an air infusion port, the other end of the high temperature and high pressure sample allocation device is connected with the constant speed and constant pressure pump, the model mechanism comprises a flowing section, a capillary tube section and a phase volume fraction tester, the flowing section is provided with a mold section, the mold section comprises a fixed clamping groove and a movable clamping groove, the movable clamping groove can be movably arranged in a scale sliding groove, flat panel grooves are formed in the scale sliding groove, a convection current channel is formed between the fixed clamping groove and the movable clamping groove, the two ends of the convection current channel are the inlet end and the outlet end, the inlet end is connected with the other end of the intermediate container and the end, where the air infusion port is formed, of the high temperature and high pressure sample allocation device, and the outlet end is sequentially connected with the capillary tube section and the phase volume fraction tester. The stratum mobility simulating device is capable of simulating fissures of different opening degrees and stratums with different roughness and the flowing of single-phase fluid or multi-phase fluid in the stratums at different flowing speed, through the arrangement of a corestone, the conditions of the stratums can be simulated in a more realistic mode, the data are accurate, and the application range is wide.

The invention discloses a material matching and equalizing device for PVC materials. A machine body is included, and scale sliding rods are fixed to the two sides of the top of the machine body; the outer surfaces of the scale sliding rods are sleeved with slide sleeves; a starting switch is arranged on one side of each sliding sleeve; a first material storing port and a second material storing port penetrate through the top of the machine body; connection plates are fixed above the two sides of the first material storing port and the second material storing port; springs are fixed to the bottoms of the connection plates; and the bottom ends of the springs are connected with the top of the machine body. The device is provided with first motors, connection rods, screen nets and material plugging blocks, the first motors are started by a worker, the first motors rotate so that the connection rods can be driven to rotate, the connection rods rotate so that the material plugging blocks canbe driven to rotate, so that raw materials intermittently enter a material mixing barrel to be stirred, and the raw materials entering the material mixing barrel fall onto the screen nets in advance;and due to rotation of the screen nets at the moment, the raw materials are uniformly scattered in the material mixing barrel, then material mixing is more uniform, and the material equalizing effectis improved.

The invention discloses a method for quickly detecting an excavation slope surface of a soil edge slope. The method is characterized in that a positioning bolt is moved in a vertical sliding groove, an angle of a detection bevel scale is adjusted, the angle of the detection bevel scale and the angle of a detection transverse scale are the same with the design angle of the soil edge slope, the detection bevel scale is parallel with the slope surface, and the position of a horizontal detection air bubble in a horizontal detection scale is used for detecting whether the excavation slope surface of the soil edge slope meets the design standard or not. A detection device used by the method comprises the detection transverse scale, wherein a vertical frame is arranged under the back end of the detection transverse scale and is vertical to the detection transverse scale; the vertical slide groove is arranged in the vertical frame; the front end of the detection transverse scale is rotationally connected with the front end of the detection bevel scale; the back end of the detection bevel scale is provided with a bevel scale sliding groove; a positioning bolt is arranged at the crossing part of the vertical sliding groove and the bevel scale sliding groove; the detection bevel scale is connected with the vertical frame through the positioning bolt; a vertical scale is arranged under the middle part of the detection transverse scale and is vertical to the detection transverse scale; the horizontal detection scale is arranged above the detection transverse scale; the horizontal detection air bubble is arranged in the horizontal detection scale. The method has the advantages that the excavation slope surface of the soil edge slope can be detected; the structure of the used device is simple, the operation is convenient, and the detection efficiency is improved.

The invention discloses a method for calculating and displaying a beam-Kriging information flow through a multi-scale sliding window, and the method comprises the steps: selecting the minimum window length and the sliding window, and calculating the beam-Kriging information flow; calculating a multi-scale information flow; carrying out reliability inspection; carrying out confidence check on the information flow obtained through calculation; and performing three-dimensional display of the multi-scale beam-Kriging information flow along with time change. The method has the beneficial effects that the beam-Kriging causality is displayed more comprehensively, and one parameter selected by experience of the method is reduced.

The invention discloses an automatic aircraft weight center of gravity measuring apparatus and a control method thereof. The apparatus herein includes a fixed front scale and a fixed rear scale. The center lines of the fixed front scale and the fixed rear scale are provided with a sliding middle scale thereon, and the fixed front scale, the fixed rear scale and the sliding middle scale form a triangle. The fixed front scale and the fixed rear scale both include a digital weighing table. Each digital weighing table is connected to a lifting structure therebelow. The digital weighing table is also provided with a laser displacement sensor therebelow. The sliding middle scale includes a digital weighing table. The digital weighing table is connected to a lifting structure therebelow. The lifting structure is disposed on a scale sliding module. The apparatus herein can automatically measure the weight center of gravity of various kinds of aircrafts, and effectively addresses great measurement errors and high measurement cost due to differences of weighing devices upon measuring the weight center of gravity of various kinds of aircrafts.

The invention provides a stratum mobility simulating device. The stratum mobility simulating device comprises a constant speed and constant pressure pump, an intermediate container, a high temperature and high pressure sample allocation device and a model mechanism, wherein one end of the intermediate container is connected with the constant speed and constant pressure pump, one end of the high temperature and high pressure sample allocation device is provided with an air infusion port, the other end of the high temperature and high pressure sample allocation device is connected with the constant speed and constant pressure pump, the model mechanism comprises a flowing section, a capillary tube section and a phase volume fraction tester, the flowing section is provided with a mold section, the mold section comprises a fixed clamping groove and a movable clamping groove, the movable clamping groove can be movably arranged in a scale sliding groove, flat panel grooves are formed in the scale sliding groove, a convection current channel is formed between the fixed clamping groove and the movable clamping groove, the two ends of the convection current channel are the inlet end and the outlet end, the inlet end is connected with the other end of the intermediate container and the end, where the air infusion port is formed, of the high temperature and high pressure sample allocation device, and the outlet end is sequentially connected with the capillary tube section and the phase volume fraction tester. The stratum mobility simulating device is capable of simulating fissures of different opening degrees and stratums with different roughness and the flowing of single-phase fluid or multi-phase fluid in the stratums at different flowing speed, through the arrangement of a corestone, the conditions of the stratums can be simulated in a more realistic mode, the data are accurate, and the application range is wide.

The invention discloses a large-scale sliding block structure for preventing a product from burring. The large-scale sliding block structure is used for forming a larger entire surface. The large-scale sliding block structure comprises a cavity insert, a core cavity and a large-scaling sliding block, wherein the core insert is combined with the cavity insert to form the product; the large-scale sliding block is arranged between the core insert and the cavity insert to form the larger entire surface of the product, and an inserting breaking parting plane is arranged between the large-scale sliding block and the core insert. Compared with the prior art, the large-scale sliding block structure for preventing the product from burring, disclosed by the invention, has the advantages that because the inserting breaking parting plane is adopted between the large-scale sliding block and the core insert, even if the large-scale sliding block bears a larger forming pressure when injection molding is performed, the large-scale sliding block retreats when being impacted, a gap generated by the inserting breaking parting plane between the large-scale sliding block and the core insert is relatively small, plastic can not enter the gap, so that the product can not form burrs.

The invention discloses a human body detection method based on sample transfer learning. The human body detection method based on sample transfer learning comprises the following steps: multi-dimensional human body features of positive samples and negative samples in a training set are extracted; a support vector machine is used for training to obtain an initial human body detector; the human body detector is divided into a plurality of small grids, wherein each small grid represents a small component, and the weight of the corresponding grid is used for representing the small component; the weight of the grids is updated; human body detection is conducted on an image to be detected in a scanning mode through a multi-scale sliding frame, and therefore a human body detection result is obtained. According to the human body detection method based on sample transfer learning, the detector can be structurally adjusted in a self-adaptive mode so as to be matched with the structure of the samples in a detection frame and reckon in matching loss into a scoring function, and therefore the human body detection method can be used for conducting structural transferring on each sample in a self-adaptive mode and processing shape change in a human body and other conditions.

The invention discloses a method for dividing a level set image based on characteristics of a neighborhood probability density function, which mainly solves the problem existing in the prior method that the variability of texture elements of a natural image in texture scale and texture direction cannot be effectively processed. The method comprises the following steps of: initializing a level setfunction; setting an initial value and a maximum value for the iteration number of the level set function; working out a characteristic set of the neighborhood probability density function; working out characteristics of a subset probability density function; working out a similarity vector; according to an evolution equation of the level set, updating the level set function; judging whether to enter a cycle, if so, working out the characteristics of the subset probability density function, the similarity vector and the level set function again; otherwise stopping the cycle, and using a zero level set of the level set function as a segmentation border of an input image. Because the texture elements of the natural image are fit by a multi-scale sliding window, the variability of the textureelements in texture scale and texture direction can be better processed, so that the method can be applied to the segmentation of the natural texture image.

A fuel tank cap with a metering function aims to solve the problems that the fuel quantity in a fuel tank can be known only through a sensor with a complex structure and the cost is high. The fuel tank cap comprises a cap body, and is characterized in that graduation marks are arranged at the upper end of the cap body; a transparent window is arranged on one side edges of the graduation marks; a scale sliding block is arranged in the scale trough arranged under the transparent window; an indication rod arranged in a screen stencil cylinder is arranged under the cap body in a position corresponding to the scale sliding block; the screen stencil cylinder is connected with the cap body; and a floating body is sleeved on the indication rod. The fuel tank cap is smart in structure, is designed reasonably, can measure fuel quantity accurately, occupies a small space, and is high in compatibility, low in manufacturing cost and strong in practicability. Therefore, requirements on fuel tank caps of various sizes can be met.