304 results about "Complex algorithm" patented technology

The present invention represents an advancement on the current processes involved in designing, engineering, modeling and manufacturing of orthotic and prosthetic devices. Orthotic and prosthetic computer aided design software has the option to apply measurements to a template to create a patient specific model. The use of algorithm generated predictions (also referred to as “AGP”) software takes this functionality and makes it more scientific. Algorithm generated predictions is the process of predicting the appropriate size and shape data through the use of complex algorithms. Certain key pieces of data are entered into the software that then calculates the appropriate dimensions and the appropriate computer aided design template. The dimensions are then applied to the computer aided design template. The computer aided design software modifies the templates by reducing and enlarging areas as necessary and a custom computer aided design model is created that can then be transformed into a physical model for the manufacture of the device.

Techniques are provided for evaluating heart failure within a patient based on ventricular impedance measurements. Briefly, values representative of ventricular end-diastolic volume (EDV) are detected using ventricular electrodes and then heart failure, if occurring within the patient, is evaluated based on ventricular EDV. In this manner, ventricular EDV is used as a proxy for ventricular end-diastolic pressure. By using ventricular EDV instead of ventricular end-diastolic pressure, heart failure is detected and evaluated without requiring sophisticated sensors or complex algorithms. Instead, ventricular EDV is easily and reliably measured using impedance signals sensed by implanted ventricular pacing/sensing electrodes. The severity of heart failure is also evaluated based on ventricular EDV values and heart failure progression is tracked based on changes, if any, in ventricular EDV values over time.

The cost of performing sophisticated software-based financial trend and pattern analysis is significantly reduced by distributing the processing power required to carry out the analysis and computational task across a large number of networked individual or cluster of computing nodes. To achieve this, the computational task is divided into a number of sub tasks. Each sub task is then executed on one of a number of processing devices to generate a multitude of solutions. The solutions are subsequently combined to generate a result for the computational task. The individuals controlling the processing devices are compensated for use of their associated processing devices. The algorithms are optionally enabled to evolve over time. Thereafter, one or more of the evolved algorithms is selected in accordance with a predefined condition.

The invention discloses a pedestrian detection method based on an end-to-end convolutional neural network in order to solve the problem that the existing pedestrian detection algorithm has the disadvantages of low detection precision, complex algorithm and difficult multi-module fusion. A novel end-to-end convolutional neural network is adopted, a training sample set with marks is constructed, and end-to-end training is performed to get a convolutional neural network model capable of predicting a pedestrian candidate box and the confidence of the corresponding box. During test, a test picture is input into a trained model, and a corresponding pedestrian detection box and the confidence thereof can be obtained. Finally, non-maximum suppression and threshold screening are performed to get an optimal pedestrian area. The invention has two advantages compared with previous inventions. First, through end-to-end training and testing, the whole model is very easy to train and test. Second, pedestrian scale and proportion problems are solved by constructing a candidate box regression network, the pyramid technology adopted in previous inventions is not needed, and a lot of computing resources are saved.

The invention discloses a method and system for carrying out searching on data. The system comprises a communication device, a cache database, a fetching server and a search server, wherein when the number of keywords matched with a search term and searched in the cache database according to a preset matching rule and the number of search results corresponding to the keywords are less than preset numbers, the obtained search results of the search server are sent to a client, and the search results of the search server are used as the supplement of the search results of the cache database. According to the method and system for carrying out searching on data, disclosed by the invention, the problem that in the prior art, when an information database and an index database are simultaneously set, a data matching process can be completed by using a complex algorithm so that the waiting time of users is too long is solved; and a beneficial effect of quickly finding matched data according to a preset cache database and a preset matching rule can be achieved.

A detecting and error-correcting method capable of rapidly and accurately obtaining an element center and a deflection angle is disclosed. Interested edge area can be rapidly and accurately found out through an automatic edge area extraction method based on combination of binary morphology and image reduction, and subsequent edge extraction image scope is minimized as possible. Pixel level edge is roughly positioned at first, and accordingly a sub-pixel edge is rapidly extracted by one-dimensional curve fitting method. The method overcomes the defect of complex algorithm of a conventional sub-pixel edge extraction algorithm based on two-dimensional images, and the time of extracting accurate sub-pixel edge can be shortened. Weighted least square rectangular edge fitting algorithm based on linear hazen paradigm can detect straight lines on any positions of an image, and effectively minimizes influence of outlier on fitting precision due to uneven edges. The central positions of a plurality of elements and deflect angles can be rapidly and accurately detected at once, and the efficiency of visual detection is improved.

The invention discloses a data search method and a data search system. The data search method comprises the following steps: pre-extracting a key word list, obtaining query results corresponding to key words in the key word list through visiting an external data source server, and associatively storing the key words and the query results corresponding to the key words in a cache database; obtaining a search request with a search word sent by a client, distributing the search request to the cache database, searching the key word matched with the search word and the query result corresponding to the key word according to a preset matching rule in the cache database; and sending the query result corresponding to the key word to the client. According to the data search method and the data search system, the problem of excessively long waiting time of a user caused by application of a complex algorithm to completion of a data matching process when an information database and an index database are configured at the same time in the prior art can be solved, and the benefit of quickly searching matched data according to the preset cache database and the matching rule can be obtained.

Stochastic models of aircraft flight paths and a method for deriving such models from recorded air traffic data. Each stochastic model involves identifying the flight plan for one or more aircraft; identifying important parameters from each flight plan, such as aircraft type, cruise altitude, and airspeed; optionally identifying flight plan amendments for each flight; representing each route of flight as a series of navigational fixes; representing at least one aircraft flight parameter probabilistically; modeling realistic differences in at least one dimension between each planned route of flight and the flight path as it might actually be flown; and communicating the modeled deviations or simulated flight paths to the user. At least one aircraft flight parameter is represented as a random variable with a particular statistical distribution, such as a normal (Gaussian), Laplacian, or logistic distribution; or with a more complex algorithm containing one or more random elements. The modeled flight parameters may be any of lateral position, longitudinal position, climb altitude, descent altitude, climb airspeed, descent airspeed, cruise airspeed, cruise altitude transition, or response time to a flight plan amendment.

The invention discloses a method for evaluating quality of streaming video without reference. The method is realized in a simulation environment of simulating network loss on MPEG-TS stream; network damage is simulated through man-made loss, simultaneously loss data is recorded, attribute of the data is analyzed, an image of the loss is stored, and impact to visual effect after the data loss is observed by eye, thereby establishing a one-to-one correspondence. In practical application, a QOE evaluation value which is in accord with subjective feeling of a user can be acquired through sampling and analyzing a transmission media and analyzing the network damage. In the method, video flowing is directly analyzed without source file information, analysis process is simple and complex algorithm is not needed, quality of user experience is reflected in real time and an accurate evaluation value is given, and an objective evaluation method and a subjective evaluation method are combined to ensure the evaluation value is closer to the real feeling of the user.

The invention relates to a handheld thermal infrared imager. The handheld thermal infrared image comprises infrared imaging sensors, a visible light camera, a laser range finder, a Bluetooth module, a magnetic compass module, an MEMS micro gyroscope, a GPS module, a wireless transmission module, a Linux system embedded signal control panel and an LED display screen. Data of the multiple sensors can be integrated, resources can be reasonably allocated, complex algorithms can be achieved, the performance of handheld equipment can be improved, and the systematization of single machines can be improved. The invention further discloses a Linux-system-based algorithm for the handheld thermal infrared to carry out quick locking and ranging on a small target. According the principle of the MEMS micro gyroscope and the image processing algorithm, the target can be quickly and accurately locked and traced through digital images of infrared light and visible light, and the quick locking and ranging functions are effectively achieved through the laser range finder. The handheld thermal infrared can eliminate video jitters in real time, improve the video quality, lock the target quickly, recognize and detect the target distance automatically and improve the functionality of the handheld equipment.

The invention discloses an intersection multiline public traffic vehicle priority request conflict coordination control method. By aiming at an intersection with the multiline public traffic vehicle priority request conflict and comprehensively considering factors such as travel time delay of vehicles, priority strategies and the road grade, a real-time multiline priority request conflict coordination control strategy is established. Only when travel time delay of one public traffic vehicle exceeds a certain threshold value, a priority request meets the limiting condition and the benefit brought by priority is higher than delay caused by the priority, the public traffic vehicle sends a priority request signal; when a multiline priority request is generated, a priority level evaluation model is utilized to sort conflicted priority requests, and the priority of passing is given to a line with the highest priority level. On one hand, the problem of unfairness of an original multiline priority request conflict solution is solved; on the other hand, parameters which the method relates to are fewer and are very easy to obtain, a complex algorithm, an optimizing process and the like are not required, practicality is high, and operating efficiency of an urban public traffic system can be greatly improved.

The present invention is directed to a system, method and software program product for implementing 3-D Complete-Body-Screening medical imaging which combines the benefits of the functional imaging capability of PET with those of the anatomical imaging capability of CT. The present invention enables execution of more complex algorithms measuring more accurately the information obtained from the collision of a photon with a detector. The present invention overcomes input and coincidence bottlenecks inherent in the prior art by implementing a massively parallel, layered architecture with separate processor stacks for handling each channel. The prior art coincidence bottleneck is overcome by limiting coincidence comparisons to those with a time stamp occurring within a predefined time window. The increased efficiency provides the bandwidth necessary for increasing the throughput even more by extending the FOV to over one meter in length and the execution of even more complex algorithms.

The invention relates to a sub module capacitance and voltage balancing and optimizing method for a modularized multi-level converter. The system used in the method is a modularized multi-level conversion system, and a flexible direct current power transmission system valve base control system framework taking a distributed parallel processing mode as a framework is established. The method comprises the following steps of (1) determining a control target of the balancing control for the sub module capacitance and voltage; (2) establishing a sub module state decision optimizing model; (3) performing balancing and equalizing on the sub module capacitance and voltage; (4) verifying the balancing and equalizing on the sub module capacitance and voltage. The distributed computation framework is adopted, the processing capability is greatly improved, and the aim of the application of a complex algorithm and a complex strategy in a high-voltage large-capacity flexible direct current power transmission system is ensured.

The embodiment of the invention discloses a method and a device for generating and receiving an OOFDM (Orthogonal Frequency Division Multiplexing) signal and a wavelength division multiplexing system, belonging to the communication field, wherein the method for generating the OOFDM signal comprises the steps of: converting one path of serial high-speed data into N paths of parallel low-speed data; respectively modulating and mapping the N paths of parallel low-speed data to obtain N paths of information sequences; expanding the N paths of information sequences into 2N+2 paths of information sequences with Hermitian symmetrical structures; carrying out rapid inverse Fourier conversion on the 2N+2 paths of information sequences to obtain an OFDM baseband signal; carrying out digital-to-analogue conversion on the OFDM baseband signal to obtain an OFDM analog signal; and modulating the OFDM analog signal to an optical carrier wave to obtain the OOFDM signal. The OOFDM signal generated in the embodiment of the invention is received without tracking and compensating phase and frequency with a complex algorithm, therefore, the cost of realizing the system is lowered.

The embodiment of the invention discloses a database function-based rule implementation and application method and a database function-based rule implementation and application device. The method comprises the following steps of: establishing a business rule model comprising one or more rule key elements according to rule application requirements; establishing a plurality of rules which conform to the business rule model; converting the rules into rule description language and saving in a rule base; implementing the rules in the rule base as corresponding database functions; configuring incidence relation of source data and the applied rule according to application requirements; and acquiring values of the key elements in the configured rule input by a user, executing the rule by calling the database function associated with the rule, and returning an execution result. By using the method or the device provided by the embodiment of the invention, the implementation of a complex algorithm is realized by means of a simple database function based on the implementation result of a rule engine implemented by the database function, the development difficulty of the rule engine is reduced, the cost is saved, and the rule processing efficiency is increased.

Active RFID technologies are used to tag assets and people within buildings or open areas, such as parking lots or military bases; and to identify, preferably within a few meters, the real-time location (RTL) of the tag, i.e. to create a real-time location system, known as an RTLS. A novel and complex algorithm is provided that combines signal strength computations with factors that incorporate physical realities, such as walls and floors, to determine a tag's real-time location more accurately.

The invention relates to an interacted system distribution type tide computing method based on computational model marriage, which comprises the following steps: defining the handshaking branch and boundary node aggregate between subnets; respectively forming subnet information amount; selecting a main network and a slave network; merging the information amount of the main network and the slave network to obtain overall information amount and consequently obtain an overall tide computation equation set; generating a node number check list at the main network side or generating respective node number check lists of the subnets; unitedly selecting a voltage iteration initial value; solving at the main network side to obtain an overall integral tide result; returning the tide data of the slave network from the main network to the slave network in a complete transmission mode or a partial transmission mode to complete the distribution type tide computation. The invention avoids a large amount of model marriage and subsequent maintenance brought about by full network integral modeling or actual model merging, avoids complex algorithm organization patterns and simultaneously obtains a computing result in full accordance with that of centralized modeling.

A method and apparatus for enhancing the receiving and information identification functions of multiple access communications systems by employing one or more optical processors configured as a bank of 1-D correlators. The present invention is particularly useful in a DS/SS CDMA communications system, resulting in a multiuser CDMA system that approaches carrier to noise performance (C/N) as opposed to being limited by multiple access interference (MAI). The correlators are arranged in parallel to detect and/or demodulate the received signal, in conjunction with one or more complex algorithms to perform near-optimum multiuser detection, perform multipath combining and/or perform carrier Doppler compensation. An improved receiver in accordance with the present invention comprises means for receiving a plurality of signals transmitted through a communications channel; signal conversion means for converting the received signals into a form suitable for input to the multichannel correlator; a multichannel optical correlator for identifying the presence of particular waveforms and estimating the relative time delay or delays, carrier frequency offset from expected, RF amplitude and RF phase for each received spread spectrum waveform present in the received plurality of signals; a controller for determining and providing to the optical correlator the appropriate set of reference hypotheses; and one or more receiver algorithms depending on the exact receiver function to be performed.

A wireless multimedia sensor network-based regional target tracking method is an application proposal based on the positioning by a wireless multimedia sensor network and is mainly used for solving the problems of node positioning and target tracking in a wireless multimedia network. In the method, a target tracking task is assigned to different nodes, and the nodes in different states and at different positions in the network cooperate to complete the target tracking task defined by a user. The invention also provides a corresponding method for detecting the regional target tracking. On the basis of a simple statistic instruction, the method can effectively detect the motion trace of a target monitored in the network in real time, avoid excessive network and node resource consumption caused by increasing network tracking precision based on a complex algorithm and ensure the objective accuracy, integrity and timeliness of the target tracking, thereby meeting the application needs of the wireless multimedia sensor network and further guaranteeing the safety of the wireless multimedia sensor network.