Abstract: Systems and techniques for performing a scan-free reading of a structured identifier are provided. One technique involves obtaining a first set of images of an object within a scanning volume of an environment. The object is passed through the scanning volume without an explicit presentation of the structured identifier to an optical scanner within the environment. For each image of a plurality of the first set of images, an indication of a first region of interest of the structured identifier on the object within the image is obtained, based in part on a second set of images of the object having a lower quality than the first set of images. The first region of interest of the structured identifier is extracted from a first image of the plurality of the first set of images. The structured identifier within the first region of interest within the first image is decoded.

Abstract: An apparatus, system and method of providing industrial analytics. The apparatus, system and method include at least a plurality of sensors sensing performance indicators for an industrial process across multiple nodes, wherein ones of the multiple nodes are remote from each other; at least one machine learning module comprising non-transitory computing code executed by a processor. When executed by the processor, the code causes the steps of: receiving user input regarding at least the industrial process and a data set; selecting a model based on at least the user input, wherein the selected model comprises a plurality of learnings based on the performance indicators sensed by multiple sensors across at least multiple ones of the multiple nodes; applying the selected model to the data set; assessing at least the performance indicators for the data set upon application of the selected model; and outputting the assessed performance indicator.

Abstract: Systems and methods are configured for correcting the orientation of an image data object subject to optical character recognition (OCR) by receiving an original image data object, generating initial machine readable text for the original image data object via OCR, generating an initial quality score for the initial machine readable text via machine-learning models, determining whether the initial quality score satisfies quality criteria, upon determining that the initial quality score does not satisfy the quality criteria, generating a plurality of rotated image data objects each comprising the original image data object rotated to a different rotational position, generating a rotated machine readable text data object for each of the plurality of rotated image data objects and generating a rotated quality score for each of the plurality of rotated machine readable text data objects, and determining that one of the plurality of rotated quality scores satisfies the quality criteria.

Abstract: The present invention relates to a system and method for reading a two-dimensional array code. The two-dimensional array code comprises a plurality of equal-height code characters arranged into a two-dimensional array, wherein each of the code characters corresponds to a bit group representing information, and the adjacent code characters in the row direction of the two-dimensional array have equal center distances; and a locating mechanism, for defining size and/or direction of the two-dimensional array arranged by the code characters. The present invention greatly reduces the cost for issuing a two-dimensional code, and significantly increases the user convenience. This two-dimensional array code has high compatibility for user mobile terminals. In addition, the entire system apparently improves its safety and stability, and is suitable for the present network application environment.

Abstract: Disclosed are systems and methods for configuring a vision detector, wherein a training image is obtained from a production line operating in continuous motion so as to provide conditions substantially identical to those that will apply during actual manufacturing and inspection of objects. A training image can be obtained without any need for a trigger signal, whether or not the vision detector might use such a signal for inspecting the objects. Further disclosed are systems and methods for testing a vision detector by selecting, storing, and displaying a limited number of images from a production run, where those images correspond to objects likely to represent incorrect decisions.

Abstract: The present invention relates to a system and method for reading a two-dimensional array code. The two-dimensional array code comprises a plurality of equal-height code characters arranged into a two-dimensional array, wherein each of the code characters corresponds to a bit group representing information, and the adjacent code characters in the row direction of the two-dimensional array have equal center distances; and a locating mechanism, for defining size and/or direction of the two-dimensional array arranged by the code characters. The present invention greatly reduces the cost for issuing a two-dimensional code, and significantly increases the user convenience. This two-dimensional array code has high compatibility for user mobile terminals. In addition, the entire system apparently improves its safety and stability, and is suitable for the present network application environment.

Abstract: Techniques are presented for analyzing audio-video segments, usually from multiple sources. A combined similarity measure is determined from text similarities and video similarities. The text and video similarities measure similarity between audio-video scenes for text and video, respectively. The combined similarity measure is then used to determine similar scenes in the audio-video segments. When the audio-video segments are from multiple audio-video sources, the similar scenes are common scenes in the audio-video segments. Similarities may be converted to or measured by distance. Distance matrices may be determined by using the similarity matrices. The text and video distance matrices are normalized before the combined similarity matrix is determined. Clustering is performed using distance values determined from the combined similarity matrix.

Abstract: A method and apparatus for encoding an image is provided. An image coding unit, including a region that deviates from a boundary of a current picture, is divided to obtain a coding unit having a smaller size than the size of the image coding unit, and encoding is performed only in a region that does not deviate from the boundary of the current picture. A method and apparatus for decoding an image encoded by the method and apparatus for encoding an image is also provided.

Abstract: A method and apparatus for encoding an image is provided. An image coding unit, including a region that deviates from a boundary of a current picture, is divided to obtain a coding unit having a smaller size than the size of the image coding unit, and encoding is performed only in a region that does not deviate from the boundary of the current picture. A method and apparatus for decoding an image encoded by the method and apparatus for encoding an image is also provided.

Abstract: A method and apparatus for encoding an image is provided. An image coding unit, including a region that deviates from a boundary of a current picture, is divided to obtain a coding unit having a smaller size than the size of the image coding unit, and encoding is performed only in a region that does not deviate from the boundary of the current picture. A method and apparatus for decoding an image encoded by the method and apparatus for encoding an image is also provided.

Abstract: First information is about respective depths of pixel coordinates within an image. Second information is about respective depths of the pixel coordinates within a ground plane. In response to comparing the first information against the second information, respective markings are generated to identify whether any one or more of the pixel coordinates within the image has significant protrusion from the ground plane. In response to a particular depth of a representative pixel coordinate within the image, a window of pixel coordinates is identified that is formed by different pixel coordinates and the representative pixel coordinate. In response to the respective markings, respective probabilities are computed for the pixel coordinates, so that the respective probability for the representative pixel coordinate is computed in response to the respective markings of all pixel coordinates within the window. In response to the respective probabilities, at least one object is detected within the image.

Abstract: An imaging technology having high reproducibility of a document and making image correction possible, without being influenced by a way of placing the document or a position of a camera, and irrespective of paper color and despite of damage of a corner portion. By printing positioning symbols on document paper in advance when the document is photographed by a digital camera or a portable phone with a camera, such as a smart phone, and by processing the image based on these positioning symbols, image processing having high reproducibility of the document and making image correction (keystone correction) possible can be implemented irrespective of paper color and despite damage to a corner portion.

Abstract: A method for counting blood cells in a sample of whole blood. The method comprises the steps of: (a) providing a sample of whole blood; (b) depositing the sample of whole blood onto a slide, e.g., a microscope slide; (c) employing a spreader to create a blood smear; (d) allowing the blood smear to dry on the slide; (e) measuring absorption or reflectance of light attributable to the hemoglobin in the red blood cells in the blood smear on the slide; (f) recording a magnified two-dimensional digital image of the area of analysis identified by the measurement in step (e) as being of suitable thickness for analysis; and (g) collecting, analyzing, and storing data from the magnified two-dimensional digital image. Optionally, steps of fixing and staining of blood cells on the slide can be employed in the method.

Abstract: An imaging technology having high reproducibility of a document and making image correction possible, without being influenced by a way of placing the document or a position of a camera, and irrespective of paper color and despite of damage of a corner portion. By printing positioning symbols on document paper in advance when the document is photographed by a digital camera or a portable phone with a camera, such as a smart phone, and by processing the image based on these positioning symbols, image processing having high reproducibility of the document and making image correction (keystone correction) possible can be implemented irrespective of paper color and despite damage to a corner portion.

Abstract: A marker generating system is characterized in having a special feature extracting element that extracts a portion, as a special feature, including a distinctive pattern in a video image not including a marker; a unique special feature selecting element that, based on the extracted special feature, selects a special feature of an image, as a unique special feature, that does not appear on the video image; and a marker generating element that generates a marker based on the unique special feature.

Abstract: The invention relates to a method for identifying sheets on which is added one identifier or more comprising a combination of juxtaposed screen cells. The method comprises two parts. The first part describes a means for making the differentiating identifiers and the addresses associated with the content areas forming the sheets. The second part describes a means for processing the information of the digital pen transmitted to the computer processing unit by one or more automatic or semiautomatic post-processing operations. The two non-separable parts of this method thus compensate for the potential handling errors of a user who must write or transcribe information onto screen sheets using a digital pen.

Abstract: In a method and apparatus for identifying an embossed character, light of one color is directed in one direction across the embossed character to illuminate certain character parts and light of another color is directed in another direction across the embossed character to illuminate other character parts. Image data for the two colors are captured and are subjected to separate image processing to detect edges highlighted by the directed light. The processed images are combined and supplemented with OCR analysis before being compared with predicted characters. Based on the comparison, a determination is made as to the probable identity of the character.

Abstract: An imaging technology having high reproducibility of a document and making image correction possible, without being influenced by a way of placing the document or a position of a camera, and irrespective of paper color and despite of damage of a corner portion. By printing positioning symbols on document paper in advance when the document is photographed by a digital camera or a portable phone with a camera, such as a smart phone, and by processing the image based on these positioning symbols, image processing having high reproducibility of the document and making image correction (keystone correction) possible can be implemented irrespective of paper color and despite damage to a corner portion.

Abstract: A branch-lane entry judging system includes a traffic-line detecting means for detecting traffic lines on right and left sides of a vehicle, a length and width detecting means for detecting lengths and widths of the traffic lines in a real space, a lane-type judging means for judging types of the traffic lines and a type of a lane in which the vehicle is running, a branch-lane detecting means for detecting a branch lane when one of the traffic lines changes from a solid line to a broken line having a predetermined length and a predetermined width, the branch lane existing near the changed traffic line, and a branch-lane entry judging means for judging that the vehicle will enter the branch lane when the vehicle moves toward the branch lane or a directional indicator is operated toward the branch lane while the branch lane is detected.

Abstract: An image processing apparatus executes acquiring, on a first image having a pattern having first areas and second areas that have a different color from the first areas, center position of the pattern where the first areas and the second areas cross, acquiring boundary positions between the first and second area, converting the first image to a second image having its image distortion corrected by using the center position and the boundary positions, acquiring, by scanning on the second image, expectation values which are areas including the point where the first and second areas cross excluding the center position, acquiring a intersection position of the intersection on the second image based on the expectation values, acquiring the center position and the positions on the first image corresponding to the intersection position by inverting the second image to the first image, determining the points corresponding to the acquired positions as features.

Abstract: A receiving system that can receive and process 3D images and a data processing method of the same are disclosed. The receiving system includes an image receiving unit and a display unit. The image receiving unit receives a 3-dimensions (3D) image and system information including additional information of the 3D image (i.e., additional 3D image information), generates 3D signaling information based upon the additional 3D image information included in the system information, and transmits the generated 3D signaling information along with the 3D image through a digital interface. And, the display unit receives the 3D signaling information along with the 3D image through the digital interface, formats the 3D image based upon the receiving 3D signaling information, and displays the formatted 3D image.

Abstract: The present invention provides an imaging technology having high reproducibility of a document and making image correction possible, without being influenced by a way of placing the document or a position of a camera, and irrespective of paper color and despite of damage of a corner portion. By printing positioning symbols on document paper in advance when the document is photographed by a digital camera or a portable phone with a camera, such as a smart phone, and by processing the image based on these positioning symbols, image processing having high reproducibility of the document and making image correction (keystone correction) possible can be implemented irrespective of paper color and despite of damage of a corner portion.

Abstract: A document processing method comprises adding document markers to predetermined locations of an electronically stored document. These are printed with the document. The document is scanned and the scanned document markers are used to process the scanned image. This processing comprises at least pixel threshold setting, and determination of the locations of the scanned image which are to be processed to derive the pixels of a digital version of the document. This enables local deformations in the paper document to be corrected, and enables correct thresholds for the printing and scanning operations to be applied. The electronically stored document can be processed to derive a set of document properties which can be used when constructing the digital version.

Abstract: An image processing apparatus includes a user interface for setting an instruction from a user and is capable of switching a language used in a display screen of the user interface. The image processing apparatus creates text data by determining the language used in the display screen of the user interface and by performing character recognition suitable for recognizing a document of the determined language on read image data. The image processing apparatus also creates a file in which the text data and the image data are associated with each other. Therefore, character recognition is properly performed in the image processing apparatus by automatically selecting the type of the language.

Abstract: After pre-printed cards are manufactured at least a corner of each card is imaged. It is then determined whether the position and orientation of markings in the image with respect to the corner are within tolerance. If not, the card is rejected. If the markings are, or include, indicia, these indicia may be identified to allow identification of the card. This allows a determination of whether the card is in an intended order, or it allows selection of the batch into which the card is placed. It also means that cards in a given set of cards arranged in a first order can be randomized by first buffering cards in buffers and then outputting the cards to an output conveyor.

Abstract: A method is provided for synchronizing corresponding landmarks among a plurality of images of an elastic object. The method includes identifying a plurality of landmarks in a first image of the object and a second image of the object, determining a correspondence between the landmarks in the first image and the landmarks in the second image, determining a distance transformation between a pair of adjacent landmarks in the first image and the corresponding pair of adjacent landmarks in the second image, and when displaying the first and second images, using the distance transformation to smoothly navigate between the adjacent landmarks such that corresponding landmarks of the first and second images are arrived at about simultaneously during navigation.

Abstract: The object of the invention is to achieve a correctly positioned print in duplex-printing mode. A method for the detection of marks (1, 1?, 1?) by a sensor array (10) for a printing machine has been provided, whereby the marks (1) on the first (recto) printing side (5) of a sheet (3) are detected, the sheet is turned over and shifted in a direction transverse to the transport direction, and the marks (1?) on a second (verso) printing side (6) of said sheet are detected. Furthermore, a printing machine is provided with an alignment device (40) for shifting a sheet (3) in a direction transverse to the transport direction for the detection of marks (1?) on the second printing side, said marks being offset relative to the marks (1) on the first printing side, after said sheet (3) has been turned over.

Abstract: An image processing system includes an image input section, a first character-extracting section, a second character-extracting section and a synthesizing section. The image input section inputs images which are linked to each other in time series manner. The first character-extracting section extracts a character from a first image input by the image input section. The second character-extracting section extracts a character from a second image input by the image input section. The synthesizing section generates a character string in accordance with the character extracted by the first character-extracting section and the character extracted by the second character-extracting section.

Abstract: Systems and computer-implemented methods for use in body pose estimation are provided. Training data is obtained, where the training data includes observation vector data and corresponding pose vector data for a plurality of images. The observation vector data is representative of the images in observation space. The pose vector data is representative of the same images in pose space. Based on the training data, a model is computed that includes parameters of mapping from the observation space to latent space, parameters of mapping from the latent space to the pose space, and parameters of the latent space. The latent space has a lower dimensionality than the observation space and the pose space.

Abstract: A device and a method are provided for spatially locating and tracking an object, such as medical instruments, a patient and/or part of a patient's body using a camera array having at least two cameras, and for mutually assigning the spatial position of the object and an acquired, stored data set. Trackable markings are recorded via the camera array, the markings are extracted from the recordings, and a location of the markings is monitored, wherein the spatial position of the object is based on the location of the markings. Further, video recordings of a surface of the object are produced via the camera array and, with computer assistance, surface portions of the object are identified and correlated with corresponding surface portions in the stored data set, wherein the spatial position of the object is assigned to that of the data set.

Abstract: A system and method for recognition of images may include the use of alignment markers. The image recognized may be a pattern from an array, a character, a number, a shape, and/or irregular shapes. The pattern may be formed by elements in an array such as an identification marking and/or a sensor array. More particularly, the system and method relate to discriminating between images by accounting for the orientation of the image. The size and/or location of alignment markers may provide information about the orientation of an image. Information about the orientation of an image may reduce false recognitions. The system and method of image recognition may be used with identification markings, biosensors, micro-fluidic arrays, and/or optical character recognition systems.

Abstract: According to a function 188 which stores in the first memory for difference, pickup image data from CCD camera 42 based on a predetermined timing, a function 190 which stores in the second memory for difference, pickup image data from CCD camera 42 based on another timing, a function 192 which obtains a difference between the pickup image data stored in the first memory for difference 24 and the pickup image data stored in the second memory for difference 26, a function 194 for specifying an image having moved based on the data as a difference, a function 196 which determines whether or not the image having moved is touching the character image, and a function 200 which increases a value of parameters such as experiential value, physical energy, and offensive power, when it is determined that the image having moved comes into contact with the character image, it is possible to expand a range of card game used to be played only in a real space up to a virtual space, and offer a new game which merges the card g

Abstract: A method and system for performing image mark recognition for a document is disclosed. A document is scanned into a digital image. Reference image marks are sensed from the digital image. The reference image marks may include trigger row marks and/or corner/crop marks. Coordinates denoting the location of cells within the digital image are determined base on the locations of the reference image marks. Response marks are evaluated for darkness, opacity, and/or grayness on a pixel-by-pixel basis. The response marks are each assigned a percentage value based on the ratio between a total color value for a cell and a maximum color value for a cell.

Abstract: An OCR percentage matching algorithm achieves a significant reduction in false mismatches accounting for combinations of unprocessed spaces, missing characters, extra characters and character substitution errors during the OCR scanning processing and allows for a specified percentage of the OCR character scan rather the entire OCR character scan to be the same as the expected character string to declare a match.

Abstract: A pressure sensitive cursor control suitable for applications that include cellular phones and personal data assistants includes a pressure sensor array suitable to detect a human fingerprint. Cursor control is provided by subtle finger movement. Sequential images of the finger are captured and compared to determine planar movement or data entry. Performance may be improved by sampling zones within the captured image and comparing the zones with those in an earlier image. Performance could also be improved by subsampling the array. While the full resolution of the pressure-sensing array is needed to capture ones fingerprint and execute an authentication function only a subsampled image from the array is needed to derive meaningful navigation data. The pressure sensor array is designed on a glass substrate in order to drive down the unit cost.

Abstract: A method and system for performing image mark recognition for a document is disclosed. A document is scanned into a digital image. Reference image marks are sensed from the digital image. The reference image marks may include trigger row marks and/or corner/crop marks. Coordinates denoting the location of cells within the digital image are determined base on the locations of the reference image marks. Response marks are evaluated for darkness, opacity, and/or grayness on a pixel-by-pixel basis. The response marks are each assigned a percentage value based on the ratio between a total color value for a cell and a maximum color value for a cell.

Abstract: Techniques for generating a paper document that integrates notes written by a user and different types of captured or recorded information. A user can take written notes during an information communication event. While taking the notes, the user can identify, from information captured during the information communication event, information that is of interest to the user and that the user would like to integrate with or incorporate into the user's notes. A paper document is generated by printing a visual representation of the notes written by the user and the visual representations of portions of information that a user would like to integrate with the user's notes.

Abstract: In a pattern inspection method, a master pattern serving as a reference and the continuous tone image of a pattern to be measured that is sensed by a camera are aligned. At least the position of a base in the continuous tone image of the pattern to be measured is detected on the basis of the master pattern. At least one threshold is set on the basis of the difference from at least the density value of the base. The continuous tone image of the pattern to be measured is binarized on the basis of the set threshold. The pattern to be measured is inspected by comparing the binarized pattern to be measured and the master pattern. A pattern inspection apparatus and alignment method are also disclosed.

Abstract: Motion picture scenes to be colorized are broken into separate elements, backgrounds/sets or motion/onscreen-action. Background and motion elements are combined separately into single frame representations of multiple frames which becomes a visual reference database that includes data for all frame offsets used later for the computer controlled application of masks within a sequence of frames. Each pixel address within the database corresponds to a mask/lookup table address within the digital frame and X, Y, Z location of subsequent frames. Masks are applied to subsequent frames of motion objects based on various differentiating image processing methods, including automated mask fitting of all masks or single masks in an entire frame, bezier and polygon tracing of selected regions with edge detected shaping and operator directed detection of subsequent regions. The gray scale actively determines the mask and corresponding color lookup that is applied in a keying fashion within regions of interest.

Abstract: A method and system for performing image mark recognition for a document is disclosed. A document is scanned into a digital image. Reference image marks are sensed from the digital image. The reference image marks may include trigger row marks and/or corner/crop marks. Coordinates denoting the location of cells within the digital image are determined based on the locations of the reference image marks. Response marks are evaluated for darkness, opacity, and/or grayness on a pixel-by-pixel basis. The response marks are each assigned a percentage value based on the ratio between a total color value for a cell and a maximum color value for a cell.

Abstract: A camera system includes a display monitor which displays an image of an object, taken by an optical unit, on a screen of the monitor. A reading unit reads a preceding image and a current image among a plurality of partially overlapping images, from a memory device, the preceding image and the current image containing a common element. A determining unit determines a positional relation between the preceding image and the current image based on a common pattern derived from the common element in the two adjacent images read by the reading unit. A displaying unit displays an image indicating a boundary of the preceding image on the screen of the monitor at a shifted position according to the positional relation determined by the determining unit, with the current image concurrently displayed on the screen of the monitor.

Abstract: A handwritten character pattern is input, a coordinate of a rightmost end or a lowermost end of the input handwritten character pattern is detected, a predetermined interval value is added to the coordinate to calculate a coordinate of a position for displaying a blank guide, and the blank guide is displayed in the same display area as that of the handwritten character pattern. In the case where a new handwritten character pattern is input in an area present on the side of the handwritten character pattern from the blank guide, it is recognized that a blank character is not present in a region on the side of the handwritten character pattern from the blank guide. In the case where a new handwritten character pattern is input in an area on the right side or the lower side from the blank guide, it is recognized that a blank character is input in an area on the handwritten character pattern from the blank guide.

Abstract: Techniques for high accuracy media positioning in a swath printer. A high accuracy media positioning method includes mounting a computer-controlled printing element for movement along a swath axis for swath printing of an image on a print medium, moving the printing element along the swath axis and printing at least a portion of a swath of the image on the print medium, sensing the position of an edge of the just printed portion of the image which is nominally aligned with the scan axis; providing relative motion between the print medium and the printing element to accurately position the printing element in dependence on the sensed position of the edge of the just printed portion of the image. The fine compensation needed to compensate positioning errors can be performed prior to printing a swath, or even “on the fly” during the printing of a swath.

Abstract: A technique which can generate stable virtual page marks even in a non-OCR form to stably identify the form. Key segments which can be stably detected within a form are predefined for the detection of virtual page marks, and the virtual page marks (circumscribed rectangle) are generated on the basis of the key segments. Redundancy is bestowed on the detection of the key segments, and substitute segments are defined so as to generate the virtual page marks on the basis of them even in a case where the key segment has failed to be detected due to a stain, a partial lack or the like.

Abstract: In a document in which a plurality of data items of different kinds are mixed, when one data item is edited, the relative positional relation to other data items is prevented from being destroyed, whereby information is prevented from becoming meaningless or from being changed. For example, when an edit is carried out on one data item, a deviation amount of that data item is derived and a shift process by the same amount is effected on the other data items, whereby the relative positional relation can be maintained among the data items.

Abstract: An encryption device includes a card with a grid comprising multiple rows and columns defining boxes for receiving individual characters. One of the columns or rows contains a master password or code character string. Other columns or rows contain encrypted character strings. The individual characters of the master password can be offset with respect to corresponding, individual characters of the encrypted character strings. A sleeve slidably, reciprocably receives the card and assists with aligning the characters of the master password and the characters of the encrypted character strings, which are visible through a window in the sleeve. A method of storing, encrypting and retrieving character strings utilizes the encryption device.

Abstract: An aberration mark for use in an optical photolithography system, and a method for estimating overlay errors and optical aberrations. The aberration mark includes an inner polygon pattern and an outer polygon pattern, wherein each of the inner and outer polygon patterns include a center, and two sets of lines and spaces having a different feature size and pitch that surround the outer polygon pattern. The aberration mark can be used to estimate overlay errors and optical aberrations. In some embodiments, the mark can also be used with scatterometry or scanning electron microscope devices. In other embodiments, the mark can be used to monitor aberrations of a lens in an optical photolithography system.

Abstract: A computer-implemented method of generating a glyph in which a plurality of design characteristics are extracted from a target font and a source glyph program is generated using the design characteristics. The source glyph program defines a new glyph that corresponds to a character that is not present in the target font, e.g., the Euro, and is stylistically similar to glyphs of the target font. The new glyph may be added to the target font, it may be rendered on an output device, or it may be saved separately from the target font.

Abstract: An expanded-image generating apparatus is provided for original image data, in which a plurality of pixels is arranged in a matrix, so as to obtain expanded-image data partitioned into a plurality of blocks, each of which is composed of a plurality of pixels When the magnifying power regarding at least one of a direction along width and a direction along length in the original image data is set, a pixel arranging processor arranges each pixel in the original image data at a position corresponding to a center position of each of the plurality of blocks in accordance with the magnifying power. The expanded-image data corresponding to the magnifying power is generated at pixel generating positions corresponding to the each block by applying a fluency transform to each pixel in the original image data at the position.

Abstract: An address reader method and apparatus for recognizing a receiver address on a surface of mail. In the invention an image of the surface of the mail is input and segmented into at least one character string candidate. At least one address area candidate is extracted from the image based upon the segmented character string candidate. One of the address area candidates extracted from the image is selected as a receiver address of the mail by analyzing each of the address area candidates based on predetermined position information indicating a usual position of a receiver address area, character direction information indicating a character direction of a character string appropriate for the predetermined position information, and key character string information indicating a character string most likely to exist in a receiver address. Characters in character strings of the selected address area candidate are recognized as a receiver address which is used to sort the mail.