JP4657532B2 - Shape transfer device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a shape transmission device which converts the shape and features of image data read by an image reader into sound data and outputs the sound data. SOLUTION: Image data is read, and outline information 2 of the image data is extracted as the shape and features of image data, and the X axis of outline information is replaced with sound position information 4 utilizing sound source orientation which is a human aural function which decides the direction of outputted sounds, and the Y axis is replaced with a frequency 5 of sounds, and image data is converted into sound data to output the sound data, and thus the shape and features of image data can be recognized by hearing.

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention is a technique for converting the shape and characteristics of image data (1) recognized and read by an image reading device into sound data and outputting the sound data.
[0002]
[Prior art]
As an output method for transmitting information of image data (1) recognized and read by a conventional image reading apparatus to a user, the input image data (1) found on an image scanner or printer is displayed on a monitor or paper. The method is used to output to and visually transmit. In addition, a method of transmitting the shape of the recognized image data (1) seen in walking for the visually handicapped person to the sense of touch by generating vibration and electrical stimulation, and the input image data (1). The method of transmitting the shape of the sound to the auditory sense using sound is used.
[0003]
[Problems to be solved by the invention]
In the conventional method, when the information of the recognized and read image data (1) is output to a monitor or paper, there are some obstacles in the field of vision and vision, including the case of operating in a place with poor visibility such as a dark place. In the state, there is a problem that it is difficult to visually recognize the image data (1). In addition, it is difficult to grasp the shape and characteristics of the read image data (1) when using a method of generating vibration or electrical stimulation and transmitting it to the tactile sense and a method of transmitting to the auditory sense using sound. There is.
[0004]
[Means for Solving the Problems]
[0005]
By extracting the contour information (2) of the image data (1) read by an image reading device such as a CCD camera or a scanner, the shape and characteristics of the image data (1) are extracted. Specifically, the horizontal direction of the image data (1) is the X axis, the vertical direction of the image data (1) is the Y axis, and the image is read in the X axis direction. As the shape and characteristics of the image data (1), The extracted contour information (2) is converted into sound data. At this time, the X-axis is replaced with the position information (4) where the sound is output, and the Y-axis is replaced with the sound frequency (5), so that the shape and characteristics of the image data (1) are output as sound data. It is possible for a visually impaired person who is difficult to recognize to recognize the shape and characteristics of the image data (1) by hearing.
[0006]
DETAILED DESCRIPTION OF THE INVENTION
Even when the object itself is two-dimensional or three-dimensional, the contour information (2) of the object can be expressed in two dimensions. In addition, the human auditory sense is called sound source localization (3), which recognizes the direction of the output sound with a function for determining the direction from which the output sound can be heard, and the sound of the output sound pressure level is detected. It recognizes the pitch difference and tone color of the sound output by the volume level and the sound frequency (5). Therefore, the horizontal direction of the image data (1) is taken as the X axis, the vertical direction of the image data (1) is taken as the Y axis, and image information is read in the X axis direction to obtain the contour information (2) of the image data (1). Extracted as the shape and features of the image data (1). By converting the extracted contour information (2) into sound data and outputting it, the shape and characteristics of the image data (1) can be recognized by hearing. At the time of sound data conversion, the X axis is replaced with position information (4) from which sound is output, and the Y axis is replaced with sound frequency (5). However, in human hearing ability, there is no recognition ability for the same frequency that is output simultaneously, and there are individual differences in recognition ability for different frequencies that are output simultaneously. Furthermore, in the sound source localization (3) for recognizing the position of the sound to be output, the ability to recognize the position of the sound output from the front and left and right is higher than the ability to recognize the position of the sound output later. By outputting the position information (4) where the sound is output from the left side in the X-axis direction in a time-scanning manner, the characteristics of the image data (1) are output as sound data. As an example, the output sound data is different at each point when n points (6), n + 1 points (7), and n + 2 points (8) have elapsed since the start of outputting image data (1) as sound data. At n point (6), two sounds of different frequencies are output simultaneously from the left side. At n + 1 point (7), two sounds having different frequencies are output simultaneously from the center, but a sound having a higher frequency and a sound having a lower frequency than those at n point (6) are output. At n + 2 points (8), no sound is output.
[0007]
In FIG. 11, after image data (1) is read from the scanner input unit, binarization processing (10) is performed by image conversion (9). The image data (1) that has been read is subjected to outline tracking processing (12) as feature extraction processing (11) to extract the features of the image data (1). Contour information (2) After image measurement (13), the table is replaced with sound data from image data (1) (14), position information (4) in which sound is output in the X direction in which image reading is performed, contour information ( Sorting (15) is performed with the vertical direction of 2) as the Y coordinate. The sorted (15) image data (1) is output as sound data.
[0008]
FIG. 12 shows an example of converting image data and specific points into sound data.
In the figure, the captured image (21) and the specific point (20) are mixed on the screen. The position of the specific point (20) is analyzed by the frequency (22) where the specific point is located and the volume (23) where the specific point is located.
(21) L. ch and (21) R.M. ch is a waveform in which the left and right output levels and the waveform when the captured image (21) is analyzed, and the frequency and the left and right output levels change depending on the shape of the captured image (21).
(20) L. chA and (20) R. The chA analyzes the position of a specific point (20), and is spaced at regular intervals so that it can be distinguished from surrounding image data by the frequency (22) where the specific point is located and the volume (23) where the specific point is located. In the example shown in FIG. chB and (20) R. chB analyzes the position of the specific point (20), determines the frequency (22) at which the specific point is located, and the horizontal direction in which the specific point (20) is located by the volume (23) at which the specific point is located. This is an example of transmission at regular intervals during the playback time.
And (21) L. ch and (20) L. chA or (20) L. chB waveform and (21) R.R. ch and (20) R.M. chA or (20) R.I. The chB waveforms are synthesized and output.
[0009]
[Brief description of the drawings]
[Fig. 1] An example of image data [Fig. 2] A diagram extracting contour information of image data [Fig. 3] A diagram showing a definition for converting image data into sound data [Fig. 4] [Fig. Is extracted and expressed as 2-axis data. [FIG. 5] The contour information of a three-dimensional figure is extracted and expressed as 2-axis data. [FIG. 6] A diagram showing sound source localization (3).
FIG. 7 is a diagram illustrating sound data output for a captured image.
FIG. 8 is a diagram showing sound data (6) to be output.
FIG. 9 is a diagram showing sound data (7) to be output.
FIG. 10 is a diagram showing sound data to be output.
11 is an example of a flow for converting image data into sound data. FIG. 12 is an example of converting image data and a specific point into sound data.
DESCRIPTION OF SYMBOLS 1 Image data 2 Contour information 3 Sound source localization 4 Sound output position information 5 Sound frequency 6 n point 7 n + 1 point 8 n + 2 point 9 Image conversion 10 Binarization process 11 Feature extraction process 12 Outline tracking process 13 Image measurement 14 Table replacement 15 Sort 20 Specific point 21 Captured image 22 Frequency at which specific point is located 23 Volume at which specific point is located

Claims (2)

A device for outputting the shape and characteristics of image data (1) recognized and read by an image reading device such as a CCD camera or a scanner as sound ,
The horizontal direction of the image data (1) is the X-axis, the vertical direction of the image data (1) is the Y-axis, the image is read in the X-axis direction, and exists at the X coordinate at the time of reading on the image data (1). Contour information (2) including the X coordinate and Y coordinate of the contour is extracted from the contour of the figure ,
The X coordinates of the contour information (2), the sound source localization (3) location of the information in the output of sound using the sound output is judged human hearing functions or heard from which direction (4) Replace the Y coordinate with the sound frequency (5) ,
By scanning the image data (1) in the X-axis direction and outputting the sound specified by the position information (4) and the sound frequency (5) corresponding to the X coordinate at the time of scanning ,
Shape transmission apparatus capable Ru is recognized by hearing the shape and characteristics of the image data (1). 2. The shape transfer device according to claim 1 , wherein a specific point is determined from the captured image, and transmission and non-transmission are repeated at regular intervals or a tone is changed according to a frequency and a volume corresponding to the coordinates of the portion. the method allows the specific point, the shape transfer device enabling that the positional relationship of the surrounding situation Ru is recognized by hearing.

Images