JP2548386B2 - License plate usage code recognition method - Google Patents

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a license plate use code recognition method for recognizing a license plate of a vehicle by imaging a traveling vehicle with a television camera and processing a still image obtained there. Is.

2. Description of the Related Art In image processing, binarizing multivalued data with gradation and compressing the data is a very commonly used method, and there are various methods.

There is a method in which all data in the processing area are added and an average value thereof is calculated and used as a binarization level. When high-speed processing is required, depending on the processing data, it takes time to perform addition operation, but this can be dealt with by thinning out the processing data to some extent.

There is also a method of creating a histogram from the data in the processing area and separating the character part and the background part. FIG. 2 shows a method of creating a histogram from character data and obtaining a binarization level. In the figure, a is processed data and b is a histogram of the data. Although there are two peaks in the histogram, the black peak is the character portion and the white peak is the background portion. From this characteristic, the binarization level (-・-
-The character part and the background part can be appropriately separated in (-line).

As described above, the conventional method can also detect the binarization level that appropriately binarizes the multivalued image data.

Problems to be Solved by the Invention However, the above conventional methods have the following problems. When the binarization level is detected using the average value, the average value is affected by the character portion ratio of each character, and therefore there is a problem in using it directly as the binarization level. For characters with a high character portion ratio, the average value is lower than the black level, and conversely, when the ratio is low, the average value is higher than the white level. If these are directly used as the binarization level, the character having a high character portion ratio is binarized to the fine character and the character having the low character portion ratio is binarized to the thick character. It is not preferable that the character shape influences the level. It is also susceptible to sunlight halation and noise.

On the other hand, when the histogram is used, there is no problem as described above, but there is a problem of how to set the gradation width when creating the histogram. The characteristics of the data cannot be clearly detected unless the gradation width is properly taken.

However, since the characteristics of the image greatly change depending on the illuminance in the image pickup state, there are cases where the characteristics of the image cannot be detected even if a histogram is created with a certain fixed gradation width.

In some images, the character portion and the background portion cannot be separated, and at this time, an appropriate binarization level cannot be derived.

The present invention solves such a conventional problem and provides an excellent method capable of detecting an appropriate binarization level.

Means for Solving the Problems In order to achieve the above object, the present invention divides characters into groups according to a character portion ratio, and binarizes the ratio for each group.

Effects According to the present invention, therefore, the ratio of the character parts is binarized so that the character parts of the binarized data and the dictionary data are matched, and the binarization level is stable against variations in the data. Has the effect of being able to detect

Practical Example FIG. 1 shows the data structure of an example of the present invention. In FIG. 1, 1 is input multi-valued data, 2
~ 4 is the binary data binarized at each character part ratio,
Reference numerals 5 to 7 are dictionary data having respective character portion ratios.

Next, the operation of the above embodiment will be described. In the above embodiment, when multi-valued data is input, the ratio of character parts
Binarize the values so that they are 31%, 40%, and 51%, and create three binary data. Of these binary data, pattern matching processing is performed for 31% of the dictionary data in the group, 40% in the dictionary data group, and 51% in the group data. To do.
Group has a small number of characters (18, i, tsu, ko etc.)
Is composed of a dictionary pattern with a character portion of 31%. A group is composed of 48 intermediate characters with a dictionary pattern of 40% character parts, and a group is composed of 48 characters with a large number of character parts (A, N, H etc.) with a dictionary pattern of 51% character parts. ing. As a result of the pattern matching, the currently processed data is binarized at an appropriate binarization level when it is matched with the dictionary group containing the character, so that it can be correctly recognized.

As described above, according to the above-described embodiment, since the processed data is binarized at the ratio of the character parts according to the dictionary,
This has an effect that stable recognition can be performed without being affected by the shape of the characters or the influence of sunlight halation.

EFFECTS OF THE INVENTION As is apparent from the above embodiment, the present invention determines the binarization level in the binarization process of multi-valued data by matching the character part ratio of the dictionary, so pattern matching with the dictionary is performed. Has the effect of increasing the accuracy of. Since it does not depend on the level of multi-valued data, it also has an effect of being strong against the influence of external illuminance change.

[Brief description of drawings]

FIG. 1 is a diagram showing a code configuration of a usage code recognition unit in an embodiment of the present invention, and FIG. 2 is a schematic explanatory diagram of conventional binarization level detection by a histogram.

Claims (1)

(57) [Claims] 1. A method of recognizing an application code of a license plate, in which the application code of a license plate is divided into groups according to the character portion ratio of characters, and the multivalued image data is binarized so as to have the ratio of the group.