Abstract: A plurality of magnetostrictive elements made of ferrite or the like are arranged in a matrix on an upper surface of a circuit board. On the periphery of each of the magnetostrictive elements, a voice coil is arranged. When current flows to the voice coils, the magnetostrictive elements themselves expand and contract in their center axis direction (height direction), and whereby ultrasonic waves (carrier waves) are emitted into the air from the magnetostrictive elements themselves. Since the magnetostrictive elements made of ferrite or the like are used, power consumption can be reduced, compared with the case of using piezoelectric vibrators. Also, the magnetostrictive elements can be made resistant to damage even if subjected to an impact due to falling or the like. Accordingly, in a parametric speaker using ultrasonic waves as carrier waves, power consumption can be reduced and vibrating elements can be made resistant to impact damage.

Abstract: An improved an audio speaker having an electrostatic membrane pump. The electrostatic membrane pump can be an electrostatic graphene membrane pump. The method of making and using the audio speaker having the electrostatic membrane pump.

Abstract: An interface for optimizing speaker sound in a wireless speaker array installation and optimization.

Abstract: An audio headset may receive a plurality of audio signals corresponding to plurality of surround sound channels. The headset may determine, via its audio processing circuitry, context and content of the audio signals. The audio processing circuitry may process the audio signals to generate a pair of stereo signals carrying one or more virtual surround channels, wherein the processing comprises automatically controlling, based on the context and the content of the audio signals, a simulated acoustic environment of the virtual surround channels.

Abstract: An audio headset receives one or more audio signals carrying one or more audio channels and processes the audio channels to generate stereo signals for output to a left and a right speaker of the audio headset. The processing determines a number of the audio channels carried in the received audio signal(s), adjusts level(s) of the audio channels based on the determined number of audio channels and/or adjusts gain and/or phase of the audio channels to control a perceived location of a listener wearing the headset relative to a source of sounds carried in the stereo signals.

Abstract: Disclosed herein, among other things, are methods and apparatuses that provide a manufacturable RF transmission line to go through the bend area of a flexible circuit to be used in a compact design, such as in a compact hearing aid design. One aspect of the present subject matter relates to using multiple inner layers of the flexible circuit to route RF transmission. By not using outer layers, the RF transmission line will be less susceptible to delamination from the polyimide dielectric layer. One aspect of the present subject matter relates to choosing copper transmission line to have dimensions that allow for narrower transmission lines with good RF return loss. The copper transmission line dimensions also allow for manufacturing in a standard process without adding extra cost.

Abstract: A hearing aid is disclosed. The hearing aid comprises a microphone adapted to receive sound signals, an amplifier configured to amplify signals received by the microphone and output means (e.g. a receiver). The hearing aid is configured to detect if speech is received by the microphone and the hearing aid is configured to provide amplification of the detected sound signals according to a non-speech mode when no speech is detected. The hearing aid is configured to provide amplification of the detected sound signals according to a speech mode when speech is detected. The amplification carried out according to the non-speech mode is different from the amplification carried out according to the speech mode. The invention also discloses a method for amplifying sound signals received by a microphone in a hearing aid.

Abstract: An apparatus is disclosed for enhancing the sound level of portable devices by providing a sound reflector between the portable device and a case. The sound reflector includes a shield which functions to redirect output from rear facing speakers toward the front. The sound reflector can be extended for redirecting sound or retracted when not in use. A case incorporating a retractable sound reflector is also disclosed.

Abstract: This invention relates to microphone accessories for use with a hand-held microphone, and to a method of using a microphone. The microphone accessory in accordance with the invention has a body comprising an attachment formation for attaching the microphone accessory to the handle of the microphone, a holding portion to be engaged by a hand of the user, a cover portion coupled to the holding portion and substantially spaced therefrom so as to define a holding zone between the holding portion and the cover portion for receipt of the hand of the user to engage the holding portion, in use, wherein the cover portion substantially shields or covers the holding zone and/or the hand of the user; and a static non-electronically variable display device and/or an electronically variable display device. The invention also relates to a wearable microphone accessory.

Abstract: A game headset receives a plurality of channels of game and/or chat audio during game play, monitors the plurality of channels, and detect one or more sounds on the channels during the monitoring. The game headset determines characteristics of the detected one or more sound and outputs one or more signals for generating a visual representation of the detected one or more sounds based on the determined characteristics. The game headset output the signals to a game console that provides the game play. The game console displays the visual representation during the game play based on the communicated visualization information. The one or more signals is a video stream of an in-game visualization. The game headset outputs the one or more signals to a second screen communication device that generates the visual representation during the game play based on the signals.

Abstract: An earphone includes a driver and a nozzle. The driver has a transducer diaphragm. The nozzle has an internal passage having an inlet opening, an outlet opening, and a wall connecting the inlet opening and the outlet opening. The wall has an inner surface within the passage. The inner surface extends from the inlet opening to the outlet opening. The outlet opening is smaller area than the inlet opening. The inner surface is configured, based on at least one tractrix, between the inlet opening and the outlet opening. The nozzle is attached to the driver with the transducer diaphragm facing the inlet opening and the inlet opening positioned nearer to the diaphragm than is the outlet opening.

Abstract: A positioning and retaining structure for an in-ear earpiece. An outer leg and an inner leg are attached to each other at an attachment end and attached to a body of the earpiece at the other end. The outer leg lies in a plane. The positioning and retaining structure have a stiffness that is greater when force is applied to the attachment end in a counterclockwise direction in the plane of the outer leg than when force is applied to the attachment end in a clockwise direction in the plane of the outer leg. The positioning and retaining structure position an earpiece associated with the earpiece in a user's ear and retains the earpiece in its position.

Abstract: When a diaphragm formed in the shape of an oval is held by a holding member, an electrostatic electroacoustic transducer which allows stable tension without a diaphragm flexing is provided. In the electrostatic electroacoustic transducer, a diaphragm 11 and a diaphragm holding member 12 are formed in the shape of an oval, and is provided with a corrector member 21 which presses inwardly from outside both side ends on a major axis of the oval diaphragm holding member, and opens and deforms outwardly a pair of opposed long sides 12a and 12b on a minor axis side of the diaphragm holding member.

Abstract: A computer implemented method for tracking a marker on a deformable surface in augmented reality (AR) applications, comprising: detecting image-key-points in a currently processed video frame of a video-captured scene; performing key-point-correspondence searching and matching the image-key-points with model-key-points are identified from an original image of the marker, comprising: calculating an key-point matching score for each image-key-point; applying a key-point matching score filter on the key-point matching scores; restricting the searching of the image-key-points in the currently processed video frame to within same mesh block determined in a previously processed video frame of the captured video frames; and applying adaptive thresholds on the key-point matching scores in determining successful matches of the image-key-points; performing motion detection of the marker in the video-captured scene and halting the application of the key-point matching score filter and suspending the restriction on the i

Abstract: Computer program product and method for determining a location and identity of an electronic rack by a remote electronic device is disclosed. The method may include capturing an image of the electronic rack with a camera attached to the remote electronic device. The method may further include determining a visual identifying trait of the captured image of the electronic rack. The method may further include comparing the visual identifying trait of the captured image of the electronic rack to a known identifying trait of a known electronic rack identified in an inventory. The method may include identifying the electronic rack based on the comparison of the visual identifying trait of the captured image to the known identified trait. The method may further include determining a location of the electronic rack by the remote electronic device. The method may further include recording the location of the identified electronic rack.

Abstract: An action recognition system recognizes driver actions by using a random forest model to classify images of the driver. A plurality of predictions is generated using the random forest model. Each prediction is generated by one of the plurality of decision trees and each prediction comprises a predicted driver action and a confidence score. The plurality of predictions is regrouped into a plurality of groups with each of the plurality of groups associated with one of the driver actions. The confidence scores are combined within each group to determine a combined score associated with each group. The driver action associated with the highest combined score is selected.

Abstract: Merchandise management is implemented by recognizing a piece of merchandise in an image on a video in real time. A piece of merchandise and m-number of first local feature which are respectively 1-dimensional to i-dimensional feature vectors are stored after being associated with each other, n-number of feature points are extracted from an image on a video captured by an imaging unit, n-number of second local feature which are respectively 1-dimensional to j-dimensional feature vectors are generated, a smaller number of dimensions of the number of dimensions i and the number of dimensions j is selected, and a recognition that the merchandise exists in the image on the video is made when it is determined that a prescribed proportion or more of the m-number of first local feature of the selected number of dimensions correspond to the n-number of second local feature of the selected number of dimensions.

Abstract: A determination method for determining reliability of a selective binding amount of a substance to be examined obtained as detection intensity of a label when a labeled substance to be examined binds to a selective binding substance fixed as a spot on a carrier includes: determining a position of the spot in image data obtained by imaging the detection intensity in the carrier and extracting a pixel group corresponding to the spot; calculating a ratio or a difference between a median value of the detection intensity of the pixel group extracted at the determining and a median value of the detection intensity of the pixel group excluding a certain top proportion of and/or a certain bottom proportion of pixels; and determining quality of the reliability based on the ratio or the difference calculated at the calculating and a certain reference value.

Abstract: In a light control system, a captured image of a cruising direction of a vehicle is acquired, and a light source is extracted from the captured image. A probability for estimating a light source to be a vehicle light source originating from a vehicle is calculated based on light source parameters for differentiating the light source. A dark section that is darker than the periphery and is present below the light source in the captured image is extracted. The probability is set to be higher for the light source of which the dark section is extracted. The light source having a probability that is a reference value set in advance or higher is estimated to be a light source of another vehicle. When the dark section that is detected as a shadow of a vehicle is detected, the probability of the light source being a vehicle light source is set to be high.

Abstract: A method for verification of alarms is disclosed. The method involves receiving an alarm signal trigger associated with an alarm signal, receiving video data from a premise associated with the alarm signal, rapidly analyzing the video data to test for the existence a significant event, and when a significant event exists, sending a representation of a segment of interest of the video data, the segment of interest being associated with the significant event, to a user.

Abstract: A computer-implemented method of tracking a target object in an object recognition system includes acquiring a plurality of images with a camera. The method further includes simultaneously tracking the target object and dynamically building environment map data from the plurality of images. The tracking of the target object includes attempting to estimate a target pose of the target object with respect to the camera based on at least one of the plurality of images and based on target map data. Next, the method determines whether the tracking of the target object with respect to the camera is successful. If not, then the method includes inferring the target pose with respect to the camera based on the dynamically built environment map data. In one aspect the method includes fusing the inferred target pose with the actual target pose even if tracking is successful to improve robustness.

Abstract: A computer-implemented method of tracking a target object in an object recognition system includes acquiring a plurality of images with a camera and simultaneously tracking the target object and dynamically building online map data from the plurality of images. Tracking of the target object is based on the online map data and the offline map data. In one aspect, tracking the target object includes enabling only one of the online map data and offline map data for tracking based on whether tracking is successful. In another aspect, tracking the target object includes fusing the online map data with the offline map data to generate a fused online model.

Abstract: Exemplary methods, apparatuses, and systems for performing object detection on a mobile device are disclosed. A reference dataset comprising a set of reference keyframes for an object captured in a plurality of different lighting environments is obtained. An image of the object in a current lighting environment is captured. Reference keyframes are grouped into respective subsets according to one or more of: a reference keyframe camera position and orientation (pose), a reference keyframe lighting environment, or a combination thereof. Feature points of the image are compared with feature points of the reference keyframes in each of the respective subsets. A candidate subset of reference keyframes from the respective subsets is selected in response to the comparing feature points. A reference keyframe from the candidate subset of reference keyframes is selected for triangulation with the image of the object.

Abstract: A scene correlation-based target system and related methods are provided. A reference image depicts a remotely-positioned object having identifiable characteristics, wherein a reference directional vector is established relative to the reference image. A target image of a general vicinity of the remotely-positioned object has an unknown directional vector, the target image having at least a portion of the identifiable characteristics. An inertial measuring unit has a scene correlation system, wherein the scene correlation system matches the portion of the identifiable characteristics of the target image with the identifiable characteristics of the reference image, wherein a slew angle between the reference image and the target image is calculated. A target image directional vector is derived from the calculated slew angle and the reference directional vector.

Abstract: A method of estimating an orientation of a body is provided. The method includes determining a reference point based on a first region of a body, calculating a translation matrix of a world coordinate system based on the reference point; determining a first vector based on the reference point and a second region of the body, calculating a first rotation matrix rotated by an angle a about a first rotation vector, which is perpendicular to the first vector and a Z-axis of the world coordinate system, as a first rotation axis, determining a second vector based on the first vector and a third region of the body, calculating a second rotation matrix rotated by an angle ? about the Z-axis of the world coordinate system as a second rotation axis, and calculating a transformation matrix based on the translation matrix, the first rotation matrix, and the second rotation matrix.

Abstract: An image such as a depth image of a scene may be received, observed, or captured by a device. A grid of voxels may then be generated based on the depth image such that the depth image may be downsampled. A background included in the grid of voxels may also be removed to isolate one or more voxels associated with a foreground object such as a human target. A location or position of one or more extremities of the isolated human target may then be determined.

Abstract: Methods and systems are described for determining eye position and/or for determining eye movement based on glints. An exemplary computer-implemented method involves: (a) causing a camera that is attached to a head-mounted display (HMD) to record a video of the eye; (b) while the video of the eye is being recorded, causing a plurality of light sources that are attached to the HMD and generally directed towards the eye to switch on and off according to a predetermined pattern, wherein the predetermined pattern is such that at least two of the light sources are switched on at any given time while the video of the eye is being recorded; (c) analyzing the video of the eye to detect controlled glints that correspond to the plurality of light sources; and (d) determining a measure of eye position based on the controlled glints.

Abstract: A method, non-transitory computer readable medium, and apparatus for side-by-side traffic location load balancing are disclosed. For example, the method receives one or more video images of a side-by-side traffic location, determines a number of cars in a first lane and a number of cars in a second lane of the side-by-side traffic location based upon the one or more video images, calculates a delta between the number of cars in the first lane and the number of cars in the second lane and recommends the first lane or the second lane based upon the delta and a respective weighting factor associated with the first lane and the second lane to provide a load balance of the cars at the side-by-side traffic location.

Abstract: An image processing method and system of an AVM system are provided. The method includes photographing, by a controller, an environment around a vehicle to generate a top view image and creating a difference count map by comparing two top view images photographed at time intervals. Partial regions in the created difference count map are extracted and an object recognizing image is generated by continuously connecting the extracted regions of the difference count map. Accurate positions and shapes of objects positioned around the vehicle may be recognized, and more accurate information regarding the objects around the vehicle may be provided to a driver.

Abstract: An image processing apparatus that, based on an image imaged by a camera installed on a car and distance to a measurement point on a peripheral object computed by a range sensor installed on the car, draws virtual three-dimensional space in which a surrounding environment around the car is reconstructed. The image processing apparatus includes: an outline computation unit configured to compute an outline of an intersection plane between a plurality of grid planes defined in a predetermined coordinate system and the peripheral object; and an image processing unit configured to draw the outline computed by the outline computation unit on a corresponding peripheral object arranged in the virtual three-dimensional space; and the plurality of grid planes are configured with planes which are perpendicular to an X-axis, a Y-axis and a Z-axis in the predetermined coordinate system, respectively.

Abstract: A parking support apparatus includes: a captured image acquisition unit acquiring a captured image from an image capturing unit which images surrounding scenery of a vehicle; a target destination section setting unit setting a target destination section which is located around the vehicle and at which the vehicle arrives, based on the image; a vertex coordinates calculation unit calculating coordinates of a vertex of the section; an intersection coordinates calculation unit calculating coordinates of the vehicle at an intersection between a parallel line parallel to the vehicle in a width direction and frame lines configuring the section in a top view when the vertex is closer to a center side of the vehicle than the parallel line; and a destination frame superposition unit superposing a destination frame onto the image by linearly connecting positions in the image corresponding to the vertex on a side separated farther from the vehicle than the parallel line among the vertexes, and the intersection.

Abstract: A system and a method of multi-user coaching are introduced herein. Motion-sensing cameras are applied to capture images, and a depth image stitching module is applied to perform a depth image stitching process on the captured images to expand the motion-sensing range, so as to establish a virtual environment for multi-user coaching. Each user can be coached individually by a one-to-multiple approach to improve his or her motions. By using the system and the method of multi-user coaching, that is, the system can only calculates on motion similarities of the users, and instructions are fed back to each user. Therefore, the system and the method described herein can be extensively applied to various products, such as a virtual gymnasium, a virtual aerobics classroom, a virtual Budokan, and so on.

Abstract: Deformability of blood cells is quantified in a short time. A blood fluidity measurement apparatus is provided with a TV camera which photographs a stream of blood in either two areas of the internal area, entrance area, and exit area of a gate, an image processing part which calculates the velocity of the blood cells contained in the blood from the image taken by the TV camera, and a deformability calculation means which calculates the deformability of the blood cells a blood fluidity from the velocity.

Abstract: Method for automatically determining a planar event such as a bedding plane from azimuthal wellbore log data (501). The image log may optionally first be filtered using a smoothing filter (502). Then, the image log is inputted to a gradient (first-derivative) filter (503). The image log is also used to find the optimal local sinusoidal trend (or orientation) on the image by determining the best sine wave at each depth among all azimuthal curves from the image using a minimum-variance method (504). Finally, bed boundaries are located based on highest contrast (506) calculated along the sinusoidal lines on the gradient image (505).

Abstract: In various embodiments, methods, systems, and computer program products for capturing and processing digital images captured by a mobile device are disclosed. The claimed algorithms are specifically configured to perform and facilitate loan application processing by capturing an image of a document using a mobile device, and analyzing the image (optionally in conjunction with additional data that may also be captured, determined, or otherwise provided to the loan application process) to determine loan-relevant information. Select loan-relevant information may be extracted, compiled, and/or analyzed to facilitate processing of the loan application. Feedback may be provided to facilitate facile application processing, e.g. by ensuring all requisite information is submitted with the loan application. Image capture and document detection are preferably performed using the mobile device, while all other functions may be performed using the mobile device, a remote server, or some combination thereof.

Abstract: An iris biometric recognition module includes technology for capturing images of an iris of an eye of a person, whether the person is moving or stationary, and whether the person is located near the iris image capture device or at a distance from the iris image capture device. The iris biometric recognition technology can perform an iris matching procedure for, e.g., authentication or identity verification purposes. The iris biometric recognition module can be incorporated into, for example, a door lock assembly and other access controlled devices, mechanisms, and systems.

Abstract: An iris biometric recognition module includes technology for capturing images of an iris of an eye of a person, whether the person is moving or stationary, and whether the person is located near the iris image capture device or at a distance from the iris image capture device. The iris biometric recognition technology can perform an iris matching procedure for, e.g., authentication or identity verification purposes. The iris biometric recognition module can be incorporated into, for example, a door lock assembly and other access controlled devices, mechanisms, and systems.

Abstract: An apparatus and a method are provided for detecting and recording, in an electronic device, evidence of a person's situation. Situation data is detected using a situation data sensor in the electronic device. Situation data can be data representing evidence of a person's situation, wherein such evidence can show a person's place in relation to another person or thing, or the state of a person being able to interact with another person or thing. Next, biometric data of the user of the electronic device is detected. Biometric data can be data that represents one or more intrinsic physical or behavioral traits of a person. Thereafter, a report is created using the situation data and the biometric data, wherein the report includes evidence of the situation of the user of the electronic device.

Abstract: A monitoring system configured to monitor a monitoring target, the monitoring system includes: a face feature quantity extraction unit configured to extract a person's face feature quantity from input images captured by first and second imaging devices; a matching unit configured to determine whether a person in the input image is a person of a face feature quantity list by comparing the face feature quantities, wherein a second face feature quantity list is a subset or an empty set of a first face feature quantity list, the matching unit matches the face feature quantity of the person in the input image captured by the first imaging device against each face feature quantity of the first face feature quantity list, and matches the face feature quantity of the person in the input image captured by the second imaging device against each face feature quantity of the second face feature quantity list.

Abstract: A face detection apparatus detecting a face from an image which is captured by an imaging unit and includes the face, includes: a position detection unit that detects a position of a face part of the face from the image; an initial state determination unit that determines a model initial state on the basis of the position of the face part; and a model fitting unit that generates a model of the face on the basis of the image by using the model initial state.

Abstract: A face detection apparatus detecting a face from an image including the face of a target person in a region imaged by an imaging unit, includes: a model fitting unit generating a first model of the face based on a first image from the imaging unit; a tracking unit fitting the first model to a second image from the imaging unit; a position detection unit detecting positions of face parts based on the second image; and a comparison unit determining whether or not a deviation occurs between the positions of the face parts in the first model fitted by the tracking unit and the positions of the face parts detected by the position detection unit. When the comparison unit determines that the deviation occurs, the model fitting unit generates a second model of the face based on the second image.

Abstract: A fingerprint scanning and image reconstruction system and method including a fingerprint scanner providing a first scan line and a second scan line separated by a line separation distance in a scanning direction. The system includes an image reconstruction module accumulating scan lines including at least the first scan line and the second scan line over a time period t. The image reconstruction module a value for decimation (t) necessary to produce a selected y axis resolution in the scanning direction based at least in part on (line count(t)/line separation distance)*a selected y resolution, where line count(t) is the number of lines accumulated in time t, and decimation(t) indicates of whether the line count(t) is greater than or less than the number of lines accumulated as a function of the time period t that will result in a selected reconstructed image y resolution in the scanning direction.

Abstract: Disclosed are various embodiments for determining a pose of a mobile device by analyzing a digital image captured by at least one imaging device to identify a plurality of regions in a fiducial marker indicative of a pose of the mobile device. A fiducial marker may comprise a circle-of-dots pattern, the circle-of-dots pattern comprising an arrangement of dots of varied sizes. The pose of the mobile device may be used to generate a three-dimensional reconstruction of an item subject to a scan via the mobile device.

Abstract: The invention provides methods and systems for reconstructing feature intensities from pixel level data. In certain embodiments, the invention uses an empirically determined transfer function to construct a theoretical estimate of pixel level data and then iteratively updates feature intensities based on a minimum multiplicative error between the pixel level data and the theoretical estimate of the pixel level data.

Abstract: Embodiments relate to finding similar coronary angiograms in a database of coronary angiograms. An aspect includes receiving angiography data for a coronary artery, processing the angiography data to identify one or more semantic features of the coronary artery, and identifying one or more nearest coronary angiograms for each of the one or more semantic features from the database of coronary angiograms. The method also includes receiving a disease attribute associated with each of the one or more nearest coronary angiograms.

Abstract: Provided is an image processing system including a server and a scanning device connected with the server through a network. The scanning device includes a measurement unit that performs multiple times of measurement required to reconstruct one image on a target to obtain a plurality of pieces of measurement data, and a data transmission unit that transmits the plurality of pieces of measurement data obtained by the measurement unit to the server. The server includes a data reception unit that receives the plurality of pieces of measurement data transmitted by the data transmission unit, and a reconstructed image creation unit that creates a reconstructed image on the basis of the plurality of pieces of measurement data received by the data reception unit.

Abstract: An imaging system (36) includes a Positron Emission Tomography (PET) scanner (38) and one or more processors (52). The Positron Emission Tomography (PET) scanner (38) which generates event data including true coincident events and scatter events, the event data includes each end point of a line of response (LOR) and an energy of each end point. The one or more processors (52) are programmed to generate (72) a plurality of activity map and attenuation map pairs based on the true coincident events, and select (76) an activity map and an attenuation map from the plurality of activity and attenuation map pairs based on the scattered events.

Abstract: Provided herein is a method for quantifying the spatial distribution of cells or sub-cellular structures comprising (a) receiving image data comprising a plurality of biomarkers, N, wherein the data represents a spatial map of the biomarkers; (b) processing the data to obtain a set of coordinates, wherein each coordinate denotes the location of a cell or sub-cellular structure represented by a biomarker or combination of biomarkers; and (c) processing the set of coordinates into a two-dimensional symmetric (2N?1)×(2N?1) or 2N×2N matrix (D). Also provided is the use of this method for assigning subjects one or more clinical characteristics, the use of this method for forming and/or testing scientific hypotheses for one or more interventions, and apparatus comprising at least one processor configured to perform the method.

Abstract: Methods and systems disclosed herein for processing sheets of sheet material include receiving a sheet in a first sheet-accepting device. The methods and systems include detecting first information of the sheet, which includes a first serial number of the sheet. The methods and systems include receiving each of the sheets in a second sheet-accepting device. The methods and systems methods and systems include detecting second information of each sheet, which includes a second serial number of such sheet. The methods and systems include determining whether the first serial number of the sheet matches the second serial number of any of the sheets. The methods and systems include generating a report in response to determining that the first serial number of the sheet does not match the second serial number of any of the sheets.

Abstract: A device and method for measuring a critical dimension of a pattern on a display substrate is disclosed. In one aspect, the device includes a region of interest (ROI) setting unit setting a region of interest in image data, determining whether the region of interest is larger than a reference region, and generating a pattern image based on the region of interest. The device also includes a design file memory storing a plurality of design patterns, a matching unit matching the pattern image to one of design patterns, and a measurement unit measuring the critical dimension of the pattern in the pattern image. The ROI setting unit selects the image data as the pattern image and outputs the pattern image to the matching unit when the region of interest is larger than the reference region.