Based on a study of the extensive literature in handprint recognition, this paper presents a survey in this challenging field. Recognition algorithms, data bases, character models, and handprint standards are examined. Achievements in the recognition of handprinted numerals, alphanumerics, Fortran, and Katakana characters are analyzed and compared. Data quality and constraints, as well as human and machine factors are also described. Characteristics, problems, and actual results on on-line recognition of handprinted characters for different applications are discussed. New emphases and directions are suggested.

Automatic recognition of handprinted characters&#8212;The state of the art

A pattern recognition system, particularly a handprint character recognition system in which electrical binary black/white "image" of one or more handprinted characters is formed and a plurality of centers of recognition (CORs) within said binary black/white images are selected as reference points for measuring the characteristic enclave of the black/white image immediately surrounding the CORs. A library of templates of said measurements around the CORs for a plurality of known exemplary character images is stored in a memory for comparison with corresponding measurements made around the CORs of images whose class is unknown to produce "template scores" proportional to the similarity of the enclaves of the known image to the enclaves measured by templates. The generic shape of a character is expressed as a "character equation" involving template scores developed on an unknown image, and each character equation is evaluated including comparing the values of such equations, and selecting the best value to determine the generic name of the unknown character.

Character and pattern recognition machine and method

Banking apparatus for reading numeric information on bank checks, drafts and like documents. An electronic black/white pixel image of the numeric information is analyzed by a system capable of reading unconstrained, constrained, printed and typed numeric characters, locating the division between dollar and cents amounts. The invention also reads overlapping touching and not touching characters. A check is fed (330) to a scanner (331) and converted to a black and white image. A courtesy amount field (CAF) on the check is located by a CAF locator (332) and is provided to a recognition system (333).

Method and apparatus for processing bank checks, drafts and like financial documents

A system extracts a set of size invariant, rotation invariant features from pixel data of a character and sends the set of features to a statistical decision tree to effectuate automatic recognition of a character. The set of extracted features includes a first group obtained by generating arrays representable as dimensions and other geometric qualities of six minimum bounding rectangles rotated about the character, including the distances between the center points of the rotated bounding rectangles. A second group of features is extracted by generating a group of arrays representing the perimeter pixel locations, radii lengths, and direction codes. The arrays of that group are "resampled" to generate corresponding 64 point arrays. The resampled array of direction codes is smoothed and incremental direction codes are computed and compared to preselected thresholds to obtain features that indicate prominent convex and concave portions of the character. A third group of features is extracted by dividing a circle bounding the character into eight rings and 24 slices and counting the number of character pixels contained in each ring and in each slice. Various Fourier transforms, autocorrelations, moment calculations, and sorting operations are performed on many of the aforementioned arrays to yield further size and orientation invariant features. The distance between the centroid of the largest hole of the character and the center of the character is computed. The ratios of the perimeters of first and second holes of the character to the outer perimeter are computed.

Method and apparatus for generating size and orientation invariant shape features

A Peltier or Seebeck element has first and second conductive members having different Seebeck coefficients. To decrease the heat conduction from one to the other end of each of the conductive members, the cross-section area at the intermediate part in the length direction is smaller than those at both ends parts. In place of the decrease of the cross-section, the shape of the cross-section of the intermediate part of each of the conductive members may be changed by dividing the intermediate part into pieces, or amorphous silicon or the like having a heat conductivity lower than those of the materials of both end parts may be used for the material of the intermediate part. In such a way, a high-performance Peltier/Seebeck element such that the difference between the temperature of the heated portion of the Peltier/Seebeck element and the opposite portion can be kept to a predetermined temperature difference for a long time and its manufacturing method are provided.

Structure of Peltier Element or Seebeck Element and Its Manufacturing Method

A pattern recognition processing system in which variations in a character, pattern or the like, especially in a handwritten one, are suppressed to extract its invariable characteristics, ensuring accurate recognition of the character, pattern or the like. The pattern to be recognized is divided into circumscribed quadrangular areas and scanned in horizontal and vertical directions to extract reflection segments between the vertical frames of the circumscribed quadrangular areas and particular segments between adjacent ones of pattern strokes. Further, endpoints of these reflection and particular segments are checked in directions across the scanning directions to extract those endpoints which are not blocked by the pattern strokes, and the corresponding positions of the co-ordinates of these endpoints on a figure frame in the directions across the scanning directions are encoded to extract characteristics of contours of the pattern to be recognized.

Pattern recognition processing system

Disclosed herein is a character recognition process which extracts white regions surrounded by strokes of character pattern information and recognizes characters by determining the directions in which said white regions are opened. The process is provided by processing apparatus, for conversion of mesh information, disposed so as to convert a white mesh point present between two black mesh points to a black mesh point during one scanning period for scanning said character pattern information in one direction. The directions of openings of the white regions are determined based on the converted character pattern information obtained by the processing apparatus.

Character recognition process and apparatus

PURPOSE:To ensure the compensation for the rotational distortion by a comparatively simple process, by extracting the point on the pattern received the rotational distortion as the 1st coordinate position information which is given from the picture memory corresponding to the three points fixed previously on the input picture. CONSTITUTION:Input pattern 2 is read by picture input device 6 and then stored in picture memory 7. With display of the contents of memory 7 on display 9, pattern 3 received the rotational distortion is displayed. Using pattern 3, points O, O, Xa, Ya, Xb, Yb, Xc and Yc are supplied via light pen 10, and then the coordinate of each point is stored in table memory 11. If grid 4 of the mXn mesh is designated previously, coordinate values x' and y' of grid 5' are calculated to be stored in address table memory 13. Values x' and y' are then supplied to address control part 8, and part 8 gives a control so that the mesh information of mXn units may be read out of the address positions corresponding to x' and y' on memory 7. This information is regarded as the information of grid 5' having coordinate valves x and y of pattern 2 to be processed.

Rotational distortion compensation processing system for input pattern

PURPOSE:To extract input information effective to the recognition of a design automation system, by giving a different label to each straight-line component of a wiring pattern to be connected electrically CONSTITUTION:The output signal of a vector figure 5 is sent to an extraction processing part 6 for direct contact straight-line components, and the processing part 6 extracts straight-line parts of a wiring pattern contacting directly with a logical circuit block pattern Its output is supplied to a processing part 7 for labeling to give a level to each segment component belonging to the wiring pattern clearly The output is supplied to a label correction processing part 8 to correct and integrate the given label Its output is then sent to a label integration processing part 9 to give a different label to each straight-line component of the wiring pattern to be connected electrically

Separation and discrimination processing system for pattern

PURPOSE:To achieve cut-out of a character by detecting the distance from each character to a plurality of characters from one frame to another frame, and characters in the vertical direction from another frame to one frame, described in a row within a band shaped frame of a given width on a document. CONSTITUTION:Each character stored in a video buffer 2 and read in at a video observing section 1 is sequentially detected toward row direction in a hystogram forming circuit 3 for the distance of characters from one frame to another frame, and from the other frame to one frame toward vertical direction, and in this detection information, points changed by a specified value or more in the said distance are detected at a changing point detection circuit 4, and the memory address and distance of changed points are stored in a memory buffer 5, and the cut-out of characters is made by a pair detecting circuit 7 in which a pair of changing points where the distance in the row direction of changed points from each frame is within a specified distance is detected as one set, based on each set of changed points. Thus, it is advantageous that the limiting frame per each character is not required.

Yoshida Masumi

Papers

Pattern recognition processing system

Character recognition process and apparatus

Rotational distortion compensation processing system for input pattern

Separation and discrimination processing system for pattern

Cut-out system of character