Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Hiro

176 results about "Feature design" patented technology

Answer Wiki. Usually design features means any aspects of a product design which anyone (most commonly the designer or the seller) would like to emphasise as particularly valuable or attractive.

Fuel cell platelet separators having coordinate features

PCT No. PCT / US95 / 13325 Sec. 371 Date Sep. 28, 1997 Sec. 102(e) Date Sep. 28, 1997 PCT Filed Oct. 10, 1995 PCT Pub. No. WO96 / 12316 PCT Pub. Date Apr. 25, 1996Fuel cell stacks comprising stacked separator / membrane electrode assembly fuel cells in which the separators comprise a series of thin sheet platelets, having individually configured serpentine micro-channel reactant gas humidification active areas and cooling fields therein. The individual platelets are stacked with coordinate features aligned in contact with adjacent platelets and bonded to form a monolithic separator. Post-bonding processing includes passivation, such as nitriding. Preferred platelet material is 4-25 mil Ti, in which the features, serpentine channels, tabs, lands, vias, manifolds and holes, are formed by chemical and laser etching, cutting, pressing or embossing, with combinations of depth and through etching preferred. The platelet manufacturing process is continuous and fast. By employing CAD based platelet design and photolithography, rapid change in feature design can accommodate a wide range of thermal management and humidification techniques. One hundred H2-O2 / PEM fuel cell stacks of this IFMT platelet design will exhibit outputs on the order of 0.75 kW / kg, some 3-6 times greater than the current graphite plate PEM stacks.
Owner:H POWER

Integrated patient room

The present embodiments provide an integrated patient room having multiple features designed to enhance the safety, satisfaction and / or outcomes for a patient and / or caregiver. In one embodiment, the integrated patient room comprises at least one bed and at least one chair. The bed and the chair can each assume a wide range of positions, and can be pre-programmed into preferred positions. Further, the bed, chair, and / or other components in the patient room may be automatically synchronized with one another to facilitate transfer of a patient. A touch screen monitor with an intuitive graphical user interface may be employed as a communication portal. The functions of the bed, chair and / or other components in the room may be accessed and / or controlled through the communication portal.
Owner:LEIB ROGER +10

Communication signal modulation mode identification method based on convolutional neural network

The invention discloses a modulation mode identification system and method based on a convolutional neural network, which solve the problems of complex feature extraction steps and low identificationrate under a low signal-to-noise ratio in the prior art. The simple feature in the identification system is constructed as a simple feature using a co-directional component and a quadrature componentof a baseband signal as signals, and the simple feature is sent to a convolutional neural network module for identification. The identification method comprises the steps of: modulating a transmittedsignal and performing pulse shaping; performing up-conversion on the transmitted signal and then transmitting the transmitted signal through an additive white Gaussian noise channel; performing pre-processing first by a receiving end to obtain the co-directional component r(t) of the analyzed signal; constructing the simple feature, i.e., constructing the co-directional component r(t) and the quadrature component of the analyzed signal into a two-dimensional matrix; performing feature learning and classification by the convolutional neural network; and sending a modulation method to a demodulation end to obtain a demodulated signal. The method is low in feature design complexity, avoids explicit feature extraction, has high classification correctness, and can be applied to communication systems having high recognition performance requirements.
Owner:XIDIAN UNIV +1

Multi-functional training garment

An absorbent article including features facilitating toilet training of a wearer is disclosed. In particular, the absorbent articles may have one or several of a variety of multi-functional training features. These features include features designed to facility ease of application and removal of the garment such as informational graphics or slow recovery elasticized features. Other features may be designed to provide training feedback to a wearer upon a urination incident or upon remaining dry for a period of time. These features may be employed in combination in a multi-functional article to provide both multi-sensory feedback and ease of application and removal features.
Owner:THE PROCTER & GAMBLE COMPANY

Polarized diffractive backlight

A backlight is provided for illuminating an at least partially transmissive display. The backlight includes a light source. A light guide receives the light from an edge surface and guides the light by total internal reflection. The light is extracted from the lightguide using sub-wavelength extraction features designed on the basis of two interleaved grating structures. The emitted light, using this arrangement has a high level of polarization.
Owner:SHARP KK

Pulmonary nodule detection device and method based on shape template matching and combining classifier

A pulmonary nodule detection device and method based on a shape template matching and combining classifier comprises an input unit, a pulmonary parenchyma region processing unit, a ROI (region of interest) extraction unit, a coarse screening unit, a feature extraction unit and a secondary detection unit. The input unit is used for inputting pulmonary CT sectional sequence images in format DICOM; the pulmonary parenchyma region processing unit is used for segmenting pulmonary parenchyma regions from the CT sectional sequence images, repairing the segmented pulmonary parenchyma regions by the boundary encoding algorithm and reconstructing the pulmonary parenchyma regions by the surface rendering algorithm after the three-dimensional observation and repairing; the ROI extraction unit is used for setting a gray level threshold and extracting the ROI according to the repaired pulmonary parenchyma regions; the coarse screening unit is used for performing coarse screening on the ROI by the pulmonary nodule morphological feature design template matching algorithm and acquiring selective pulmonary nodule regions; the feature extraction unit is used for extracting various feature parameters as sample sets for the post detection according to selective nodule gray levels and morphological features; the secondary detection unit is used for performing secondary detection on the selective nodule regions through a vector machine classifier and acquiring the final detection result.
Owner:KANGDA INTERCONTINENTAL MEDICAL EQUIP CO LTD

Bimodal emotion recognition method and system based on 3D convolutional neural networks

The invention discloses a bimodal emotion recognition method and system based on 3D convolutional neural networks. According to the method, first, two 3D convolutional neural networks for face emotionrecognition and body gesture emotion recognition are constructed respectively, and network model parameters are optimized based on a training set and a verification set of a bimodal face and body gesture database; second, the two neural networks obtained after optimization are tested based on a test set of the bimodal face and body gesture database to obtain a face emotion recognition confusion matrix and a body gesture emotion recognition confusion matrix; and last, priori knowledge of the face emotion recognition confusion matrix and the body gesture emotion recognition confusion matrix isutilized to fuse bimodal recognition results of a newly input face video sequence and body gesture video sequence, and a bimodal emotion classification result is obtained. Through the method, the 3D convolutional neural networks and a bimodal fusion algorithm are adopted, the subjectivity of manual feature design is avoided, the limitation of single-modal emotion recognition is overcome, and the accuracy and robustness of emotion recognition can be effectively improved.
Owner:NANJING UNIV OF POSTS & TELECOMM

Arc resistant electrical enclosure

Electrical enclosures are provided that include arc resistant features designed to add structural strength for arc containment, to inhibit arc propagation, and / or to direct the release of pressure within and / or from the enclosure in order to provide arc resistant electrical enclosures. In general, the arc resistant features may be designed to provide enclosures where in the event of an arc fault, the doors and covers remain closed, parts are not ejected from the enclosure, holes are not produced in the enclosure, indicators located in close proximity to the enclosure do not ignite, and / or grounding connections remain effective. Further, the foregoing features may be designed to provide electrical enclosures that comply with industry guides and standards for arc resistant ratings.
Owner:ROCKWELL AUTOMATION TECH

Multi-mission payload system

A missile, such as a cruise missile, has a nose payload portion having a frangible nose cover and a relatively hard target penetration nose cone. The nose cone may have a liquid fuel tank within, and a chemical energy explosive charge, such as a shaped charge, aft of the liquid fuel tank. The target penetration nose cone enables perforation of certain types of targets prior to detonation of the chemical energy explosive and the liquid fuel. The frangible nose cover is configured to be easily perforated or otherwise removed by the explosive force of the chemical energy explosive charge when the missile system is utilized for the attack of hard targets. The nose payload portion may have a fragmentation case, with one or more features designed to enhance fragmentation during detonation of the explosive and / or the liquid fuel.
Owner:RAYTHEON CO

Multipurpose growing system

A multipurpose growing system for production of commercial plants which include a unique plant carrier, which allows plants to be grown and supported without any growing media, a total flow push rod and conveyor system providing automated spacing of the plants and automated disinfecting of plant carriers and troughs. The troughs feature designs for growing either with plant carriers or without them allowing various types of crops to be grown. The troughs also feature ridges extending along the bottom to limit liquid gradients and to promote aeration of the roots. The system also consists of an nutrient controlling method for hyperaccumulating of edible crops with minerals and vitamins. Finally, the system is characterized by an automatic climate control system that features an improved design making environmental control more economical by reducing air volume subject to heating and cooling.
Owner:RIPATTI MATTI T

Learning behavior feature-based SPOC (small private online course) student weekly performance prediction method and device

The present invention discloses a learning behavior feature-based SPOC (small private online course) student weekly performance prediction method and device and belongs to the online learning field. Log data in the online courses of students are collected; the learning behavior features of the students are extracted from the log data so as to train a plurality of data mining models; and a data mining model of which the performance is optimal on a training set is adopted to predict the weekly performance of the students. Correspondingly, the prediction device of the present invention comprises a data acquisition module, a feature extraction module, a training data generation module and a prediction module. According to the learning behavior feature-based SPOC student weekly performance prediction method and device of the invention, compared with ordinary learning behavior features, the learning behavior features designed based on the learning habits of the students are integrated with the teaching experience of teachers, and therefore, the learning habits of the students in the online courses can be embodied, the improvement of the prediction accuracy of the prediction model can be facilitated, and the teachers can timely know students who have troubles in learning so as to adjust the difficulty of the courses and provide targeted counseling.
Owner:BEIHANG UNIV

Pitch-based subresolution assist feature design

A method of designing a mask for imaging an integrated circuit (IC) design layout is provided to efficiently configure subresolution assist features (SRAFs) corresponding to an optimally configured annular illumination source of a lithographic projection system. A critical pitch is identified for the IC design, and optimal inner and outer radial coordinates of an annular illumination source are determined so that the resulting image projected through the mask will be optimized for the full range of pitches in the design layout. A relationship is provided for determining an optimal inner radius and outer radius for the annular illumination source. The number and placement of SRAFs are added to the mask design so that the resulting range of pitches substantially correspond to the critical pitch. The method of configuring SRAFs so that the image will have optimal characteristics, such as good contrast and good depth of focus, is fast.
Owner:GLOBALFOUNDRIES INC

Human body recognition method based on multi-base radar micro-Doppler and convolutional neural network

ActiveCN108872984AEcho Signal Disparity MitigationEnhanced recognition robustnessRadio wave reradiation/reflectionHuman bodyMultistatic radar
The invention relates to a human body recognition method based on multi-base radar micro-Doppler and a convolutional neural network, and belongs to the technical field of radar target recognition. According to the method, the multi-base radar is used so that the echo signal difference caused by the change of view can be alleviated, the recognition robustness can be enhanced and the recognition accuracy can be enhanced. The convolutional neural network is applied to perform data processing without manual feature design so as to have certain universality and excellent recognition accuracy performance. According to the method, the transfer learning technology is used, the RGB optical image is utilized in the convolutional neural network to pre-train the weight and the three-channel multi-resolution time-frequency graph having the similar RGB optical image is used as the input of the convolutional neural network so that the pre-training weight dimension is matched, more information is provided in comparison with that of the single-resolution time-frequency graph. Great recognition accuracy can be obtained in multiple human body recognition tasks by the method.
Owner:TSINGHUA UNIV

A method for recognizing human behavior in a video

The invention relates to the field of artificial intelligence, more specifically, to a method for recognizing human behavior in a video. The main algorithm idea of the invention is as follows: firstly, the input video segment is pretreated, a six-parameter affine model is used to simulate the motion of a camera, and the motion trajectory is obtained through the affine model; Secondly, the video key frames are extracted based on temporal saliency and spatial saliency. Then the improved dense trajectory is extracted from the key frames, and the two-stream convolution neural network is selected as the feature extractor to extract the depth learning feature from the key frames. The extracted features are normalized, and the features are fused based on the video representation, which fuses thevideo representation based on the depth feature and the video representation based on the improved dense trajectory feature. The invention integrates the improved dense trajectory feature IDT and thedepth learning feature designed manually, and can more effectively mine the complementary information of the two features and the behavior pattern in the video, so as to obtain better effect in videohuman body behavior identification.
Owner:SUN YAT SEN UNIV

System for providing an enhanced immersive display environment

An immersive dome includes a number of novel features designed to enhance the performance of the immersive dome over presently-known immersive dome environments. Projectors are mounted in a multi-tier tower, out of sight beneath a viewing platform positioned to provide optimal wrap-around viewing. The projection surface consists of open-cell foam that allows passage of behind-surface sound into the dome while allowing unwanted ambient noise within the dome to escape. A visually-reflective coating, in conjunction with the open cell structure, provides a textured surface that acts as a micro-baffle and suppresses cross-reflection of projected imagery.
Owner:APPLIED MINDS

Knee prosthesis

A knee prosthesis for use in a total knee replacement surgical procedure may include a femoral component, a tibial component and a meniscal component. Optionally, the prosthesis may also include a patellar component. The femoral component may include a bone attachment side and a joint facing side, the latter including an anterior joint surface, a posterior joint surface having a cross-sectional shape defining a portion of a cylinder, and medial and lateral grooves between the anterior and posterior joint surfaces. The meniscal component may include a number of features designed to mate with the femoral component to provide relatively natural movement and range of motion about the knee joint as well as stability and resistance to wear and tear.
Owner:INTPROP CHALET

Welding defect feature extraction and welding quality analysis method based on image processing

The present invention discloses a welding defect feature extraction and welding quality analysis method based on image processing. The method of the present invention comprises the following steps: S1, performing image enhancement on a grayscale images acquired by a black and white camera; S2, according to the type of workpiece and the type of the welding region, designing a workpiece background segmentation card, segmenting the background of the enhanced image, and eliminating the influence of the background on subsequent image processing; and S3, according to a feature design extraction algorithm of welding holes, obtaining shape and area information of the welding defects, analyzing the sizes of the welding holes, and automatically classifying the failure degrees of the welding holes. According to the method disclosed by the present invention, image processing technologies such as image enhancement, background segmentation, binarization processing, contour extraction, and the like are successfully applied to the actual welding scene, the welding defect features in the workpiece after welding are effectively extracted, and the defect area is calculated; and the welding quality can be automatically analyzed in real time, and improvement of factory production efficiency can be facilitated.
Owner:SOUTHEAST UNIV

Network encrypted traffic recognition method and device

The invention discloses a network encrypted traffic recognition method and device. The method comprises a preprocessing stage and a classification stage. In the preprocessing stage, original flow is subjected to flow segmentation, sampling, vectorization and standardization, a sampling scheme in large flow is provided, and the classification problem of the large flow is solved. In the classification stage, spatial feature capture and abstract feature extraction are performed by using a CNN, and then traffic time sequence features are learned by using stacked bidirectional LSTM on the basis ofabstract features so that automatic feature extraction and efficient recognition of encrypted traffic can be realized. The method has universality, can automatically extract the space-time features ofthe encrypted traffic without manual feature design of experts, and can adapt to traffic feature changes caused by different encryption technologies and confusion technologies.
Owner:NANJING UNIV OF POSTS & TELECOMM

Training method of bidirectional LSTM (long short term memory) model for implementing locomotive energy-efficient operation

ActiveCN106844949AImprove state representation abilityImproved gear prediction abilityDesign optimisation/simulationSpecial data processing applicationsData setTraining phase
The invention provides a training method of a bidirectional LSTM (long short term memory) model for implementing locomotive energy-efficient operation; the method comprises the steps of acquiring driver historical driving data and locomotive operation monitoring log as initial training data; preprocessing the initial training data to obtain a training dataset and a test dataset; in case of initial training, initializing parameters of a model directly; otherwise, continuing to train a bidirectional LSTM neural network model based on the previous bidirectional LSTM model, and saving the trained model; using the trained model to perform simulation testing on the test dataset to obtain new initial training data; using the new initial training data and the training data of the previous training phase jointly as initial training data; repeating the steps until the model converges. The feature design method, model design method and iterative training method provided herein allow data information to be fully utilized, and gear prediction capacity of the model is improved.
Owner:TSINGHUA UNIV

Design method of surgical scan templates and improved treatment planning

In implant dentistry, scan templates, also known as scan appliances or radiographic guides, are fabricated according to patients' anatomy, and later on transferred into surgical guides through CT scan, image segmentation, surface reconstruction, CAD modification, and manufacturing. This approach may be also applied to other kinds of surgery. The final manufactured surgical guides can hardly match the actual geometry of the initial scan templates, thus cannot fit well into patients' mouths, which can pose serious risks for surgery.This invention introduces calibration features to design and make a scan template, and accordingly the workflow to perform treatment planning with the template. Calibration features are geometric form features or their patterns. Iterations are utilized to make sure the reconstructed model of a scan template can replicate or match with the calibration features. After treatment planning, calibration features are removed from the model (an optional step), and a surgical guide is designed and fabricated. This invention also includes some special calibration feature designs. The treatment planning workflow is greatly improved with calibration features introduced.
Owner:GAO FEI
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products