JPH03203000A - Automatic road sign recognizing device - Google Patents

Abstract

PURPOSE:To reduce the burden of a driver and to improve safety by judging the contents of a road sign by processing a picture, which image is picked up by a television camera, and transmitting the contents to the driver. CONSTITUTION:A television camera 1 always picks up an image in front of a car and this picture signal is converted to digital data. The data are stored in a frame memory 12 as they are and on the other hand, an edge detection processing is executed by edge detection part 13. The result of the detection is sent to a feature encoding part 14 and the size of a graphic or a character string part is encoded and extracted as the feature of picture data. Next, a sign extraction part 15 retrieves a part, which feature is extracted, and executes the DMA transfer of the picture data near such a part from the frame memory 12 and the exact position of the sign is detected. Then, a sign recognition part 16 decides the contents of the sign by pattern matching and recognizes the meaning of the sign. The result of this recognition is sent to an entire control part 17 and the contents of the sign are sent out to a sound generator 3. Thus, the safety of driving can be improved.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to an automatic road sign recognition device that captures an image in front of a traveling vehicle using a television camera, recognizes the type of road sign through image processing, and informs the driver of the type of road sign. be.

Conventional technologyAs the electronics of automobile equipment progresses,
Several systems have been developed and are being considered to reduce the burden on drivers. One of these is highway radio, which provides information to cars wirelessly.

This is to inform drivers of road information.
By installing many of these radio transmitters on the road,
More detailed information can be conveyed to the driver. However, this only provides information to the driver, and there is currently no system that processes and makes decisions on the driver's information. In the future, as artificial intelligence (AI) technology becomes more sophisticated, computer-controlled autonomous driving may become a reality.
Current technology has not reached the point where it is put into practical use. On the other hand, in railways, automatic train stopping devices (
ATS), automatic train control system (ATC), etc. Using computer technology to manage the operation of transportation systems in this way will reduce the burden on drivers and greatly contribute to improving safety.

Problems that the invention aims to solveHowever, even if systemized things like railways can be controlled by computers, in order to perform similar control on a car-by-car basis, it is necessary to form a more complex network, which is currently not possible. is difficult. The present invention aims to assist the driver and reduce the burden on the driver by inputting the same information that the driver can obtain visually through a television camera and extracting the information through computer processing. The purpose is to

Means for Solving the Problems In order to achieve the above object, the present invention comprises a television camera that captures an image in front of the vehicle, an image processing device, and a sound generation device that conveys the content of the sign to the driver. The image processing device includes an analog/digital (hereinafter referred to as A/D) converter that converts the image signal captured by the television camera into digital data, a frame memory that stores the data, and a frame memory that stores the data. An edge detection unit that performs edge detection processing, a feature encoding unit that detects features from the edge information that has been subjected to edge detection processing and compresses and encodes them, a label extraction unit that detects the position of the sign, and recognizes the content of the sign. The device is equipped with a sign recognition section.

Therefore, according to the present invention, the content of a road sign is determined by processing an image captured by a television camera, and the content is communicated to the driver, so that the driver can understand the content of the sign only visually. It also has the effect of being able to be confirmed not only by hearing, but also by hearing. This has the effect of reducing the burden on the driver and further improving safety.

Embodiment FIG. 1 shows the configuration of an embodiment of the present invention. In FIG. 1, 1 is a television camera that images the front of the vehicle; 2 is an image processing device that stores and processes image signals captured by the television camera to recognize the content of road signs; and 3 is the content of the signs. This is a voice generator that conveys information to the driver.

FIG. 2 is a block diagram of the image processing device 2. As shown in FIG. In FIG. 2, 11 is an A/D converter that converts the video signal from the television camera 1 into digital data, 12 is a frame memory that stores one screen of digital data, and 13 is an A/D converter that converts the video signal from the television camera 1 into digital data. 14 is a feature encoding unit that extracts features from the edge information subjected to edge processing by the edge detection unit 13 and compressed and encoded; 15 is a feature encoding unit that performs compression encoding on 16 is a sign recognition unit that recognizes the content of the sign based on the information from the sign extraction unit 15, and 17 is an A/D conversion unit. Part 1
This is an overall control unit that controls a series of operations of the sign recognition unit 1 to sign recognition unit 16 and sends an operation command to the sound generation device 3 according to the result.

Next, the operation of the above embodiment will be explained. In the above embodiment, the television camera 1 always images the front of the vehicle, and this image signal is converted into 8-bit 2 by the A/D converter 11.
It is converted into digital data with 56 gradations. While the digital data is stored as is in the frame memory 12, it is subjected to edge detection processing by the edge detection section 13. The detection results are sent to the feature encoding unit 14 as edge information, and the shape (circle, rectangle, triangle), size, or character string part of the figure is encoded from the edge information and extracted as a feature of the image data. Next, the sign extracting unit 15 uses this encoded data to search for a portion where the characteristics of the pattern as a road sign are extracted, and transfers the image data in this vicinity from the frame memory 12 by DMA to extract an accurate sign. Detect location. In this way, when a road sign is detected from the image data, the sign recognition unit 16 determines the content of the sign by pattern matching and recognizes the meaning of the sign in the image. Regarding pattern matching, since the shapes of figures can be classified into several categories at the stage where features have been extracted in advance, pattern matching is performed only with patterns of signs belonging to those categories. This recognition result is
Sent to 7. The above is the flow of processing for one screen. If no matching pattern is found even after pattern matching is performed in the sign recognition unit 16, it is assumed that no sign exists and the process is terminated, and the next image is captured. here,
If the characteristic of the sign exists, the sign extraction unit 15 determines the distance from the current location to the sign based on the ratio of the size of the sign to the image, since the size of the sign itself is approximately constant. Then, the sign recognition unit 16 determines the content of the sign. This result is sent to the overall control unit 17, which sends the content of the sign to the voice generator 3 when transmitting the information to the driver. The audio generator 3 records audio data in advance according to the received content and outputs the RO.
Data indicating the target sign is read from M, and voice synthesis is performed to generate a voice and convey it to the driver.

As described above, according to the above-described embodiment, since the content of the road sign is automatically recognized and communicated to the driver, it is possible to support the driver's driving awareness and reduce the burden on the driver.

Effects of the Invention As is clear from the above embodiments, the present invention automatically recognizes road signs and communicates them to the driver, and has the effect of reducing the burden of driving on the driver. This results in
This has the effect of improving driving safety.

[Brief explanation of drawings]

FIG. 1 is a block diagram of an automatic road sign recognition device in one embodiment of the present invention, and FIG. 2 is a schematic block diagram of an image processing device in the same embodiment. 191. Television camera, 2. ,,image processing device,
300. Sound generator, 11. ,, A/D conversion section, 12
．． ,, frame memory, 13° edge detection section, 14
．． ,,feature encoding unit, 1511. Label extraction unit, 16. ,
, Sign Recognition Department, 1711. Overall control section.

Claims (1)

[Claims] It is composed of a television camera that captures an image in front of the vehicle, an image processing device, and a voice generating device that performs voice synthesis using voice data recorded in a ROM, and the image processing device is configured to capture images taken by the television camera. an analog-to-digital converter that converts the image signal into digital data; a frame memory that stores the data; and an edge detector that performs edge detection processing on the data;
a feature encoding unit that detects features from edge information that has been subjected to edge detection processing and compresses and encodes them; a label extraction unit that detects the position of a sign from the digital data stored in the frame memory; What is claimed is: 1. An automatic road sign recognition device, comprising: a sign recognition unit that recognizes the sign; the content of the recognized sign is uttered by the voice generating device.