2407results about How to "Improve the detection rate" patented technology

An intrusion detection arrangement (101) for communication networks comprising a network activity observer (102) configured to monitor network traffic by the related traffic elements, such as data packets, thereof and to establish traffic profiles relative to the monitored traffic elements, such as one profile per each monitored traffic element, a misuse detector (104) configured to determine a first indication of a probability of the profiled traffic representing malicious activity through co-operation with a model repository (106) comprising at least one model characterizing a known intrusion attack, an anomaly detector (108) configured to determine, at least logically in parallel with the misuse detector, a second indication of a probability of the profiled traffic representing anomalous activity through cooperation with a model repository (110) comprising at least one model characterizing legitimate network activity, and a classifier (112) configured to operate on said first and second indications to generate a classification decision on the nature of the profiled traffic, wherein the applied classification space includes at least one class for legitimate traffic and at least one other class for other traffic such as malicious and / or anomalous traffic. A corresponding method is presented.

A method of detecting objects from terrestrial based mobile mapping data is disclosed, wherein the terrestrial based mobile mapping data has been captured by way of a terrestrial based mobile mapping vehicle driving on a road having a driving direction, the mobile mapping data including laser scanner data, source images obtained by at least one camera and position and orientation data of the vehicle, wherein the laser scanner data includes laser points, each laser point having associated position and orientation data, and each source image comprises associated position and orientation data. In at least one embodiment, the method includes: retrieving a position and orientation of the vehicle; filtering the laser scanner data in dependence of the position and orientation of the vehicle to obtain laser points corresponding to regions of interest; retrieving a source image associated with the position and orientation of the vehicle; mapping the laser points corresponding to regions of interest to image coordinates of the source image to generate a recognition mask; combining the recognition mask and the source image to obtain candidate 3D images representative of possible objects within the regions of interest; and, detecting a group of objects from the candidate 3D images. By combining image recognition and laser scanner recognition the detection rate can be increased to a very high percentage, thereby substantially reducing human effort. Furthermore, the generating of regions of interest in the laser data, enables a significant reduction of the processing power and / or the processing time needed to detect the objects in the images.

The invention discloses a small target rapid detection method based on a deep convolution neural network. The deep convolution neural network is improved by the following steps: selecting the sliding windows on the convolution feature map of the last shared convolution layer of a VGG16 network as candidate boxes, wherein the sliding windows adopted are half-pixel precision sliding window; deleting a fifth pooling layer, and retaining other convolution layers and pooling layers; adding a convolution layer with a 3*3 convolution kernel; and using two convolution layers with 1*1 convolution kernels to replace all full-connection layers in the network to get the network adopted in the invention, training the network using collected data to get a small target classification model, and using the model to detect small targets. By using the method, the computational complexity is reduced, and the detection rate of small targets is improved.

A method for detecting an object in a digital image including the steps of performing color quantization on a model image including the object and on a search image that potentially includes the object, generating a plurality of search windows, computing spatial-color joint probability functions of each model and search image, where the color co-occurrence edge histogram is chosen to be the spatial-color joint probability function, assessing the similarity of each search image to the model, and designating search windows as containing the target object.

An image pickup apparatus which increases the rate at which a person's face is detected in electronic zoom and accurately focuses on a person's face area and provides exposure / white balance control. An A / D converting unit converts an analog signal obtained from an image pickup device into a digital image signal. A signal processing unit performs signal processing on the digital image signal as image data. A display unit displays a zoom area in the image data magnified / reduced to a desired angle of view. A face detecting unit detects a person's face area from image data of a face detecting area narrower than the whole area of the image data and wider than the zoom area to acquire information on the person's face area.

Scanning rate in a capacitance change measuring circuit for a capacitive sensor device is enhanced. The circuit includes: an electrode drive section line-sequentially applying an input pulse signal to a plurality of columns of first electrode patterns in the capacitive sensor device; a peak hold circuit storing a peak level of a detection signal extracted from each column of second electrode patterns into a capacitive element as a corresponding potential; a current source initializing the potential in the capacitive element within one period of the input pulse signal; a comparator comparing the potential in the capacitive element with a reference value; and determination sections each determining whether or not an input operation using a human body or the like is executed, based on a timing information and a reference timing information, the timing information representing a timing when the potential held in the capacitive element crosses the reference value.

There is described a distributed optical fibre sensor for detecting one or more physical parameters indicative of an environmental influence on a sensor optical fibre, as a function of position along the sensor fibre. The sensor uses probe light pulses of different wavelengths. At least some of the probe light pulses may also be of different pulse lengths. The relative phase bias between interferometric signals in backscattered probe light of different wavelength pulses may also be controlled.

An efficient voice activity detection method and system suitable for real-time operation in low SNR (signal-to-noise) environments corrupted by non-Gaussian non-stationary background noise. The method utilizes rank order statistics to generate a binary voice detection output based on deviations between a short-term energy magnitude signal and a short-term noise reference signal. The method does not require voice-free training periods to track the background noise nor is it susceptible to rapid changes in overall noise level making it very robust. In addition a long-term adaptation mechanism is applied to reject harmonic or tonal interference.

A method and apparatus for employing honeypot systems to identify potential malware containing messages whereby a decoy system to receive illegitimate e-mails is established. E-mails sent to the spam e-mail honeypot decoy are initially scanned / filtered and e-mails that are not considered possible malware containing e-mails are filtered out while the remaining e-mails sent to the spam e-mail honeypot decoy are identified as potential malware containing e-mails. One or more features, and / or feature values, of the identified e-mails are then identified, extracted and ranked. Once a given feature, and / or feature value, occurs more than a burst threshold number of times, the status of the given feature, and / or feature value, is transformed to that of suspicious e-mail parameter.

The invention discloses a method for realizing automatic recognition of atrial fibrillation on mini dynamic electrocardiogram monitoring equipment. The method includes: calling an MIT-BIH arrhythmia database, an MIT-BIH normal sinus rhythm database and a long-time atrial fibrillation database as training samples, and introducing an artificial neural network for learning training; setting a weight value for each layer of the artificial neural network, and inputting the training data samples to repeatedly and iteratively correct the weight value of each layer until training errors are less than a certain specified value, through the means, a weight value matrix capable of judging occurrence of atrial fibrillation can be found; utilizing the weight value matrix, and adding the same into an original artificial neural network to build a new artificial neural network; using collected electrocardiosignals of a target human body, processing the electrocardiosignals of the human body, acquiring a target human body feature vector X, and performing prediction operation according to the target human body feature vector X and the new artificial neural network.

A data discriminating unit includes a data discriminator and a sync signal discriminator. A data sync signal is detected for the bit string discriminated from the output of the sync signal discriminator, and the data is demodulated for the bit string discriminated from the output of the data discriminator, thereby optimizing the signal to a form suitable for discrimination and detection of the data sync signal and demodulation for data reproduction. As a result, the detection performance of the data sync signal can be improved while maintaining the data discrimination performance. A Viterbi decoder is used for discrimination by the data discriminator and the sync signal discriminator for the optimization thereby to optimize the number of states, the number of paths for state transition and the path memory length. A signal having no continuous inversions of the recording data is used as the data sync signal.