137results about How to "Small amount of memory" patented technology

A system and method for the capture and storage of data relating to the movements of objects, in a specified area and enables this data to be displayed in a graphically meaningful and useful manner. Video data is collected and video metadata is generated relating to objects (persons) appearing in the video data and their movements over time. The movements of the objects are then analyzed to detect the movements within a region of interest. This detection of movement allows a user, such as a manager of a store, to make informed decisions as to the infrastructure and operation of the store. One detection method relates to the number of people that are present in a region of interest for a specified time period. A second detection method relates to the number of people that remain or dwell in a particular area for a particular time period. A third detection method determines the flow of people and the direction they take within a region of interest. A fourth detection method relates to the number of people that enter a certain area by crossing a virtual line, a tripwire.

Set top boxes are provided with access to full EIT data streams without having to off-tune from one transport to another. If off-tuning is required, however, a locking mechanism substantially reduces or eliminates service interruption. Data in the additional EIT data streams in the non-home transports is reorganized to be sent by time rather than by service. EIT data is stored in the set top box in an architecture which significantly reduces the redundancy inherent in EIT data, allowing a full 7 days of EIT data to be stored in the set top box memory. The set top box of the present invention is backward compatible with conventional set top boxes.

An innovative routing method for an integrated circuit design layout. The layout can include design netlists and library cells. A multiple-level global routing can generate topological wire for each net. An area oriented graph-based detail routing on the design can be performed. A post route optimization after the detail routing can be performed to further improve the routing quality. Some methods can be single threaded all or some of the time, and / or multi-threaded some or all of the time.

An image processing device clips a plurality of partial images from an image, and calculates difference values of pixel values between a target pixel and surrounding pixels surrounding the target pixel in the partial images. Then, the image processing device decides parameter values corresponding to respective surrounding pixels based on identification information that identifies the partial image from among the plurality of the partial images and the difference values, and calculates a feature amount corresponding to the target pixel using the parameter value decided. Then, the image processing device obtains a feature amount of the partial images by calculating the feature amount with the above-described steps, using each pixel in the partial images as the target pixels.

The invention relates to a security authentication system, in particular to a security authentication system which can carry out stage treatment to the login password of a user and endows each stage password with specific functional meaning; simultaneously, the invention also relates to a set and a using method of the security authentication system. As the invention needs the user to carry out or guide the operation on site, more opportunities and time are provided for the users of the invention to get rid of the control or alarm when in a non-voluntary moment. A service end provides a personalized response code and a check code for each user; the users can firstly judge and then input the response code and the check code before inputting the passwords, thereby avoiding a fake service end.

This data encoding apparatus includes: a predictor for calculating a prediction value of target data; a subtractor for calculating the difference between the target data and the prediction value as a prediction error value; an absolute value calculator for calculating the absolute value of the prediction error value; a flatness detector for detecting the number of the absolute prediction error values continuously equal to or lower than a predetermined threshold for some thresholds, and setting a code length candidate for each threshold by comparing the detected number with a predetermined run length; a k parameter determinator for determining the code length of a fixed length code part on the basis of the code length candidate of each threshold; and a Golomb-Rice encoder for encoding the absolute prediction error value in accordance with the determined code length by dividing it into a variable length and the fixed length code parts.

A computer program product for converting from a first floating point format to a second floating point format, each floating point format having an associated base value and being represented by a significand value and a exponent value, comprising an executable algorithm to perform the steps of: determining the second exponent value by multiplying the first exponent value by a predefined constant and taking the integer portion of the result, the predefined constant being substantially equivalent to the logarithm of the first base value divided by the logarithm of the second base value; determining a bias value substantially equivalent to the second base value raised to the second exponent value divided by the first base value raised to the first exponent value; and determining the second significand value by multiplying the first significand value by the bias value.

A mail server extracts a character unregistered in a portable terminal from received mail data and affixes the font data of the character concerned to the mail data or inserts a reading tag indicating the reading (pronunciation) of the character concerned into the mail data. The portable terminal additionally registers the font data affixed to the mail data into a font database before the received mail data are displayed. Furthermore, in the display processing of the mail data, a character for which the corresponding font data is unregistered is replaced by a no-font symbol and then displayed. Furthermore, the font of each character constituting the reading tag is read out and this font is displayed subsequently to the no-font symbol.

A mail server extracts a character unregistered in a portable terminal from received mail data and affixes the font data of the character concerned to the mail data or inserts a reading tag indicating the reading (pronunciation) of the character concerned into the mail data. The portable terminal additionally registers the font data affixed to the mail data into a font database before the received mail data are displayed. Furthermore, in the display processing of the mail data, a character for which the corresponding font data is unregistered is replaced by a no-font symbol and then displayed. Furthermore, the font of each character constituting the reading tag is read out and this font is displayed subsequently to the no-font symbol.

A computer program product for converting from a first floating point format to a second floating point format, each floating point format having an associated base value and being represented by a significand value and a exponent value, comprising an executable algorithm to perform the steps of: determining the second exponent value by multiplying the first exponent value by a predefined constant and taking the integer portion of the result, the predefined constant being substantially equivalent to the logarithm of the first base value divided by the logarithm of the second base value; determining a bias value substantially equivalent to the second base value raised to the second exponent value divided by the first base value raised to the first exponent value; and determining the second significand value by multiplying the first significand value by the bias value.

A speech recognition method using a domain ontology includes: constructing domain ontology DB; forming a speech recognition grammar using the formed domain ontology DB; extracting a feature vector from a speech signal; modeling the speech signal using an acoustic model. The method performs speech recognition by using the acoustic model, the speech recognition dictionary and the speech recognition grammar on the basis of the feature vector.

A system and method for the capture and storage of data relating to the movements of objects, in a specified area and enables this data to be displayed in a graphically meaningful and useful manner. Video data is collected and video metadata is generated relating to objects (persons) appearing in the video data and their movements over time. The movements of the objects are then analyzed to detect the movements within a region of interest. This detection of movement allows a user, such as a manager of a store, to make informed decisions as to the infrastructure and operation of the store. One detection method relates to the number of people that are present in a region of interest for a specified time period. A second detection method relates to the number of people that remain or dwell in a particular area for a particular time period. A third detection method determines the flow of people and the direction they take within a region of interest. A fourth detection method relates to the number of people that enter a certain area by crossing a virtual line, a tripwire.

An innovative routing method for an integrated circuit design layout. The layout can include design netlists and library cells. A multiple-level global routing can generate topological wire for each net. An area oriented graph-based detail routing on the design can be performed. A post route optimization after the detail routing can be performed to further improve the routing quality. Some methods can be single threaded all or some of the time, and / or multi-threaded some or all of the time.

The invention discloses a safe and efficient method for preparing an emulsion matrix, which comprises the following steps: (1) preparing aqueous phase solution and oil phase solution respectively according to a process matching; and (2) conveying the aqueous phase solution and the oil phase solution into a static mixer to be mixed, preliminarily emulsifying the mixture, then conveying the mixture to an emulsifier to be finely emulsified, and obtaining the emulsion matrix after the fine emulsification is finished. The method is mainly characterized in that by connecting the static mixer in series at the front end of the mechanical stirring emulsifier, the preparation capacity of the emulsion matrix of a finished explosive production line can be improved to 3.5 to 7 t / h from 2 to 4 t / h, the preparation capacity of the emulsion matrix for mixing and charging emulsive explosive on site can be improved to 5 to 12 t / h from 3 to 5 t / h, simultaneously the production safety and the quality of the emulsion matrix are ensured, and the method is suitable for popularization and application in the technical transformation for improving the capacity of the emulsive explosive.

The embodiment of the invention discloses a voice command input method based on operation gestures. The method comprises the steps that a gesture command representing that a user selects a first application is obtained; when the gesture command is a voice service gesture command for activating the first application, the voice receiving function of the first application is started according to the input gesture command; voice input by the user and corresponding to the first application is received; the voice is identified and a first operation object to be executed is determined; according to the first operation object, a first function corresponding to the first application is executed. The embodiment of the invention further discloses an electronic device. The memory capacitance of a memory command of the user is reduced, voice service operation of the electronic device is more practical, and the user experience effect is better.

Routing methods for an integrated circuit design layout are disclosed. The layout can include design netlists and library cells. A multiple-level global routing can generate topological wire for each net. An area oriented graph-based detail routing on the design can be performed. A post route optimization after the detail routing can be performed to further improve the routing quality. Some methods can be single threaded all or some of the time, and / or multi-threaded some or all of the time.

The invention discloses a switching method of television channels and a client side for realizing switch of the television channels. The method comprises the following steps: in the channel switching process of every time, when user keying information is detected, the client side screens out a channel containing the user keying information as a new current alternative channel from current alternative channels according to a character matching scheme and displays the information of the new current alternative channel to a user; after the fact that one channel in the current alternative channels is chosen is detected, a television is switched to the channel; and during first screening, the client side uses all channels in a channel list obtained from a server as the current alternative channels. The channel list at least comprises channel numbers, channel names and channel letter abbreviations of all channels, and the information of the current alternative channels at least comprises the channel numbers and the channel names. According to the method and the client side, the memory capacitance of the user is decreased, and the error rate of the user switching the channels is obviously reduced.

Provided is a method of calculating a compensation factor to compensate for lens shading due to the characteristics of an image capturing device, which requires a small amount of memory. A reference image is captured, and a compensation factor is calculated using the characteristics of a lens shading pattern of the captured reference image. A distribution of pixel values is approximated using an exponential spline function, and a compensation factor is calculated using the approximated distribution. In addition, a method and an apparatus for compensating for lens shading by using a calculated compensation factor are provided.

The method includes 25 codes arranged on 5X5 matrix. Strokes for each code in horizontal direction are arranged in horizontal including horizontal rising, vertical, left-falling, point (right-falling) sequence. Strokes for each code in portrait are arranged in restriction of character structure from small to large. Based on position of row and column in the matrix, two digits are assigned to each code. The two digits are name of code, and value of the code. Large value means large restriction. Value of the code can be input from large keyboard or small keypad. Length of character code is 4. For example, for left and right shaped character, character codes are chose from first stroke in left part, first and last strokes in right part (largest possible code) and one code chose from other part.

An image processing apparatus includes a histogram calculation unit configured to calculate a histogram of a luminance value for each spatial block obtained by dividing an input image in a spatial direction, a gray scale correction coefficient calculation unit configured to calculate a gray scale correction coefficient used for correction of the gray scale of a luminance value for each bin of the histogram, and a gray scale correction unit configured to perform gray scale correction of a luminance value of a target pixel based on the gray scale correction coefficient of the bin to which the target pixel of the input image belongs and the gray scale correction coefficient of a bin adjacent to the bin to which the target pixel belongs in the spatial direction and a luminance direction.

The invention discloses a Chinese character computer input method, in particular a method for inputting Chinese characters of pictographic and ideographic classification. Radicals composing the Chinese characters are grouped according to certain rules and then arranged on English letter keys of a universal standard keyboard of a computer respectively by counting and analyzing a large amount of font structures of the Chinese characters and considering the use condition and character forming efficiency of Chinese character sides, Chinese character radicals and certain basic Chinese character components according to the pictographic or ideographic characteristics of the radicals. The Chinese character radicals are all divided into 26 groups, a memory supporting formula is compiled according to the names of the groups, and the main radical in each group of radicals corresponds to a key position on the keyboard. The method can allow people to reduce the memory quantity and learn in short time and make the people unforgettable for long time once mastering, and is convenient to use. The method can be widely used in computer input of the Chinese characters.

The utility model discloses an elementary vocal-based phonetic input proposal for Chinese characters, so that the Chinese characters input to a computer becomes more effective, normative and easier to use. The technical proposal is that the Chinese character vocal-based phonetic input method comprises: (1) if initials are needed, the initials of the Chinese characters can be input by pressing keys representing the initials of the Chinese characters; (2) if auxiliary finals exists, the auxiliary finals of the Chinese characters can be input by pressing keys indicating the auxiliary finals of the Chinese characters; (3) elementary vocal finals of the Chinese characters can be input by pressing keys representing the vocal finals of the Chinese characters; wherein, nine keys are chosen from the keys showing the single-letter initials and the key '';'' for inputting nine elementary multi-letter finals respectively. The proposal is applied in the Chinese character computer-input field.

A pseudorandom number generator reduced in size while maintaining high security is disclosed. The generator has a state storage unit of 2 blocks (n bits per block) and a buffer of plurality of blocks, and mixes their contents to obtain a random number sequence. The mixing is done by a nonlinear transformation unit for inputting the stored content of the buffer to output data of the same size as the input data, a first linear transformation unit for inputting the content of the state storage unit and the output of the nonlinear transformation unit to store an output into the state storage unit, and a second linear transformation unit for inputting the stored content of the buffer and the stored content of the state storage unit to store an output into the buffer. The mixed content of the state storage unit is output as a random number sequence.

The present invention concerns a method and device for coding or decoding a structured document, for example an XML document, using a structure-describing file of the XML Schema type.The method comprises a step of configuring an encoder or decoder including the obtaining of data describing a document structure model and their storage in the form of recordings,and a step of calculating a number of items, referred to as “possible items”, able to succeed, according to the structure model, an item of said model, said number of possible items being stored in a recording associated with said item.In order to facilitate calculations, the recordings comprise the minimum and maximum numbers of occurrences of the item defined by the structure model.

A system, wireless device (402) and method receive a set of soft metrics (614) associated with a block of symbols during a predetermined time interval, use the set of soft metrics (614) to form an initial interference power estimate, form a moving-average estimate of interference power (616) corresponding to each predetermined time interval, and scale the set of soft metrics (614) according to the corresponding interference power estimate resulting in a set of scaled metrics (618). Then they limit the dynamic range of the set of scaled metrics (618), linearly quantize the limited metrics, and output the linearly quantized limited metrics (628) to a memory buffer. The size of the memory buffer containing the linearly quantized limited metrics (628) is smaller than the size of a memory buffer required to store the received soft metrics (614). The linearly quantized limited metrics (628) are then used as input to a decoder.

The invented coordinate numeral method for inputting Chinese character defines 10 kinds of strokes and 10 groups of parts as well as 10 relevant digital keys. The Chinese character is split into the parts and strokes. Based on the number of the strokes of the part and the shape feature of the stroke, the relevant number key is pushed to input the Chinese character.

The input method includes the position code of the character pattern, the pronunciation code of the full character, the pronunciation code of partial character (or radical) in the full character as well as the stroke position code. The English letter keys are divided into left, right, top, bottom and unique zones. The position code of the character pattern is determined from the first stroke and the character structure. 37 radicals needed to memorize are arranged in columns of horizontal, left falling, point, vertical, turning strokes according to the first stroke. 23 initial consonants are arranged (including vowels and initial consonants such as 'ch', 'zh' etc.) The horizontal, left falling, point, vertical, turning strokes are arranged on V, U, I, A, O keys. Comparing with other input methods, the invented method has the features without need of knowing the character to be input, easy to remember, low coincident code rate and easy to split characters.

The invention relates to a font code and digital code general Chinese character input method re resolving project which is characterized in that it divides the stroke into ten strokes and the part into ten parts, then it uses them as core to design four font code and digital code general Chinese character input methods and uses the four input methods to form an integrated general Chinese character input method re resolving project of the denominator keyboard and the digital keyboard.

An image processing device clips a plurality of partial images from an image, and calculates difference values of pixel values between a target pixel and surrounding pixels surrounding the target pixel in the partial images. Then, the image processing device decides parameter values corresponding to respective surrounding pixels based on identification information that identifies the partial image from among the plurality of the partial images and the difference values, and calculates a feature amount corresponding to the target pixel using the parameter value decided. Then, the image processing device obtains a feature amount of the partial images by calculating the feature amount with the above-described steps, using each pixel in the partial images as the target pixels.