JPH06119480A - Bar-code reader - Google Patents

Abstract

PURPOSE: To provide a bar code reader which can read a bar code print even if the print does not meet standards, or is deformed or slightly separated from a bar code. CONSTITUTION: This reader is provided with a correcting means 5 which compares respective bars and space width of a bar code read signal with reference values, digit by digit, and corrects the bars and space width, digit by digit, by compression or expansion according to the rate of the distortion. The respective bars, etc., are sectioned by the size through a bar width comparison part 61 and decoded by table comparison. When the size belongs to the border area of a reference bar, a decoded value determining means 65 makes a comparison between both adjacent sections by using a conversion table 64 and only when only one value is determined, the value is determined as a decoded value. Further, the size of pulse width is regarded as input data of a membership function and a combination of various pulse width is used as a fuzzy rule for fuzzy inference, and a decoding process is performed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a bar code reader for reading a bar code in which data is coded by a combination of bars and spaces having different widths, and particularly, decoding processing is performed by using a conversion table or by fuzzy inference. The present invention relates to a bar code reading device having an improved bar code reading rate.

[0002]

2. Description of the Related Art In a conventional bar code reading apparatus, the pulse width of a bar or space following the bar width is measured with reference to the pulse width of the start bar of the bar code, and the read bar code is sequentially decoded in digit units. It was Further, the pulse width of the bar reading pulse due to the unevenness of the bar code / labeled medium is taken into consideration, and the pulse width of the reference bar is sequentially corrected with the digit width read immediately before as the reference. Further, in the conventional bar code reading device, the bar type is uniformly determined and decoded depending on whether or not the bar code width is within a predetermined range.

[0003]

However, in the prior art, first the first digit is read based on the width of the start bar, and then the reference bar width is corrected based on the first digit. When the width was less than the predetermined width, it was determined that the reading was unsuccessful to prevent misreading. Therefore, if the width of the start bar is less than or equal to the predetermined value, it cannot be read. In addition, if the bar type is uniformly decided depending on whether the bar width is within a predetermined area, if the bar code printing is out of the standard, the bar code label will be attached to the curved product. In such cases, the bar code reading rate is extremely reduced, and in some cases, the bar code cannot be read.
In particular, handheld scanners have a high probability of being unreadable unless the reading part of the scanner is brought very close to the bar code. It is an object of the present invention to provide a bar code reading device which corrects a bar code regardless of the bar width of the start bar. Further, according to the present invention, the bar code printing is slightly deviated from the standard, and the bar code label is attached to a curved product such as a fresh food product wrapped in a bottle, a can, a bag or a wrap, and the bar code is greatly distorted. It is an object of the present invention to provide a high-performance bar code reader capable of reading even if it can be read.
Another object of the present invention is to provide a bar code reading device capable of reading a bar code without degrading the reading accuracy even when the reading part of the bar code scanner is far away from the bar code.

[0004]

According to the present invention, the width (length) of the read bar or space is set to a predetermined digit reference value without using the read start bar width as a reference. Correction is performed by compressing or expanding each digit in proportion, and the corrected signal is decoded. In the present invention, the width (length) of the read bar or space is divided according to a predetermined standard, and the combination of each divided bar is decoded by the conversion table. At that time, when the width of the bar or space is near the boundary of the domain of the partition, the decoding process is attempted when the bar or space is partitioned for each of the two partitions adjacent to the boundary. Boundary data processing means is provided for processing as if the correct decoding is performed only when only one decoding value is determined. Further, in the present invention, the corrected width (length) of the bar or space is used as the input data of the membership function, and the combination of the width (length) of the bar or space for each digit of the barcode is used as the fuzzy rule to perform fuzzy inference. To perform decoding processing.

[0005]

Embodiments of the present invention will now be described with reference to the drawings. FIG. 1 is a block diagram showing an outline of the entire configuration of the barcode reading apparatus of the present invention. Light is emitted from the light source (not shown) to the barcode 1, and the reflected light from the barcode 1 is input to the data input unit 2. The data input section 2 is provided with a light sensing means (not shown), and converts the received reflected light into an electric signal. The output from the data input section 2 is sent to the signal processing means 3, where it is amplified and converted into digital signal data. Since the processing from the reflected light to conversion into digital signal data is a conventional technique and can be easily performed by those skilled in the art, further description will not be given in this respect. The pulse width of the digital signal data is then measured for each pulse by the measuring means. The measured width of each pulse is then corrected by the correcting means 5 for each digit, and then decoded by the decoding means 6. The pulse width measurement, correction, and decoding will be described in detail below. In the following description, the bar code uses the JAN code. Also,
According to the JAN code standard, the left half of the bar code is composed of a combination of an odd parity code and an even parity code, and the right half is composed of an even parity code. Therefore, in the JAN code, even if the same character is represented, the bar code pattern should be different between the first digit, the second digit, and the third digit. However, in the present embodiment, for convenience of description, unlike the original JAN code standard, it is assumed that the same code (odd parity code) is used from the first digit to the third digit.

(1) Measurement of pulse width The read signal converted into a digital signal is measured by the measuring unit 4 in FIG.
Thus, the pulse width is measured for each pulse and stored in the data storage unit 55 in units of each digit. A timer or a counter may be used for the measurement, but the measuring unit 4 of the present embodiment measures the pulse width and the digit length by the counter operating at a predetermined speed.
Therefore, the measured value is represented by the number of counts. FIG. 2 shows a waveform in which a part (first digit) of the bar code is read and digitally converted and the state of the count value in which the length thereof is measured by the measuring unit 4. The bar code in Fig. 2 is a JAN code and consists of black and white patterns of the same width.
The bar portion, the first digit portion, etc. are shown. Furthermore, each digit of the bar code consists of four black and white bars, and the width of each bar is a module (basic unit width).
It is any size from the width of one piece to the width of four pieces. Such black and white bars are called blocks in sequence. The first digit of the bar code in Figure 2 is a white bar (block) of 3 module width, each 1
It is composed of a black bar and a white bar (block) having a module width and a black bar (block) having a module width. This represents the number "9" in the odd parity representation of the JAN code standard. Below the bar code in FIG. 2, the digital waveform corresponding to the bar code is shown, and below that, the count value of each block corresponding to the width of the first digit digital pulse is shown. . The count value representing each pulse width thus measured is stored in the data storage unit 55 in units of digits as data representing the width of each bar, as illustrated in FIG.

(2) Correction The correction means 5, as shown in FIG.
2, a comparison unit 53, a correction unit 54, a data storage unit 55 and a correction data storage unit 56. Reference data storage unit 52
Stores the reference digit length (reference length) defined by the JAN code standard. Comparison unit 53
Calculates the ratio between the stored reference length and the digit length of the read digital signal. The correction unit 54
The reciprocal of the calculated ratio is multiplied by the read digit length to correct the read digit length so as to approach the reference length.
Store in 6.

The correction processing operation will be described more specifically with reference to FIGS. FIG. 4 is a flow chart showing the processing procedure of the correction processing operation, and the correction processing operation is performed in digit units. FIG. 5 shows a part of the bar code and the data before and after reading the bar code, and FIG. 6 shows a state in which the corrected data is stored in the correction data storage unit 56. In FIG. 5 and FIG. 6, the first digit to the third digit of the bar code are all shown as the numeral "9". In the following description, the letter S and the number in parentheses indicate the step number in FIG. 4, and the count value (N
N in n) represents a count number, and n represents a block number.

When the measurement and storage of the digital signal data is completed, the comparison unit 53 reads the reference digit length (L) of the standard bar code from the reference data storage unit 52 (S1).
Next, the digit length (TN = N1 + N2 + N3 + N4) is calculated by summing up the count values of blocks, and the ratio p of the measured digit length to the reference digit length (L) is p.
(= TN / L) is calculated (S2). After that, the measured count value of each block is multiplied by the reciprocal of the ratio p, that is, the correction ratio k (= L / TN), so that the digit length of the read digital signal approaches the length of the reference digit. The count value (Nn) of is corrected to obtain the standardized count value (Xn) (S
3). Here, X represents the corrected count number. The standardized count value after the correction process is the correction data storage unit 5
6 is stored. Similarly, all the digits of the read bar code are corrected and stored for each digit (S
4).

FIG. 5 shows a specific example of the correction. FIG. 5 shows an example of the correction process when the reference digit length L = 70. In the example of the first digit, the count value TN of the measurement digit is the same as the reference digit length "70", the correction ratio k is 1, and no correction is made. In the case of the second digit example, since the count value TN of the measured digit is "66", the correction ratio k is 1.06.
It becomes 1. Therefore, each block of measurement digits
The respective count values are multiplied by 1.061 to obtain standardized correction data. In the third digit example, the measured digit count value TN is the same value as in the second digit example, so the correction ratio k is 1.061 which is the same as in the second digit example, but the block different. The correction ratio k for the count value of each block
Is multiplied by and the count value of each block is corrected. Figure 6
Shows the storage state of the correction data of the first to third digits of FIG. 5 stored for each block of each digit in the correction data storage unit 56.

The reason why the correction is performed on a digit-by-digit basis is as follows. That is, when the barcode label is attached to the uneven surface having many irregularities, the distortion of each part of the barcode is also uneven. Therefore, if the entire bar code is corrected, the ratio between the bars cannot be accurately reproduced. However, since 1 digit is composed of 4 bars and its length is short (2-4 mm), even if the bar code is attached to the surface that is deformed unevenly, looking at the digit units, The distortion can be seen as uniform. Therefore, by correcting in digit units, the optically detected bar code can be corrected to the size of the reference bar code quite accurately without changing the width ratio between the bars. On the other hand, if the correction is performed for each bar, the correction process becomes complicated. Therefore, we decided to correct in digit units.

(3) Decoding processing The correction data is read and decoded by the decoding means 6. The decoding means 6 includes a bar width comparison unit 61 that compares the correction data with the reference bar width, a reference bar width storage unit 62 that stores the reference bar width, a boundary bar processing unit 63, a table 64, and a decode value determination unit 65. . The decoding means 6 has converted the correction data into the reference bar code size quite accurately by the correction described above. However, the width of each bar according to the corrected data may still be out of the standard due to defective printing of the barcode or a problem in optical reading. In such a case, if it is processed as unreadable, the reading rate is reduced accordingly.
Therefore, according to the present invention, even in such a case, the decoding means described below is used to read the information within the predetermined reference.

First Embodiment First, a first embodiment of the decoding means of the present invention will be described. First, the bar width comparison unit 61 reads the corrected bar width for each block from the correction data storage unit 56 and compares it with the reference bar width stored in the reference bar width storage unit 62. In the JAN code, there are four types of bar widths from one module width to four module widths, and the reference bar is defined for these four types and stored in the reference bar width storage unit 62. Hereinafter, these four types of reference bars are represented by M1 to M4. The width of the reference bars M1 to M4 has a certain width in consideration of the allowable range according to the standard, variations in printing of each bar, variations in optical reading, and the like. In addition, in the present invention, boundary areas E1 to E3 are provided in the boundary areas of the reference bars M1 to M4. The definition areas of the reference bar and the boundary area are illustrated in FIG. 7 (hereinafter, the bars corresponding to the boundary area are referred to as boundary bars E1 to E3). The bar width read from the correction data storage unit 56 is compared with the reference bar width to determine which type of bar M1 to M4 each read bar is for each block. When the correction data is any of the boundary bars E1 to E3 belonging to the boundary area of the reference bar, the boundary bar processing unit 63 allocates the corresponding E1 to E3. FIG. 8 shows the result of comparison for the example shown in FIG. Thus, the correction data is represented by the bar width comparison unit 61 as a combination of the reference bar and the boundary bar.

Next, the decoding confirmation processing in the decoding confirmation unit 65 will be described. The combination of the reference bar and the boundary bar from the bar width comparison unit 61 is the decoded value determination unit 6
5 is compared with a predetermined table 64. As a result of the comparison, if any of the tables is applicable, the table 64 is decoded. The conversion table 64 is created so as to correspond to a combination of bars by numbers of a predetermined standard (JAN code in this embodiment). FIG. 9 shows an example of the conversion table. As described above, in the JAN code standard, the left half of the bar code is a combination of the odd parity code and the even parity code, and the right half is the even parity code. Therefore, the conversion table corresponding to each code is Although necessary, only the odd parity code conversion table 64 is shown in this embodiment. Further, in this embodiment, for convenience of description, unlike the original JAN code standard, an odd parity code from the first digit to the third digit is used.

Now, the first digit is the reference bar M3M1M.
As it is composed of 1M2, the number from the conversion table "
The second digit has a border bar E2 with the block. If there is a border bar, replace the block with each of the reference bars adjacent to that border bar and use one of the conversion tables 64. In this example, since the block and the block are boundary bars E2, respectively, the block and
Is replaced with M2 or M3 which is a reference bar adjacent to, and the comparison with the conversion table 64 is performed. In this case, since there are two boundary bars E2, there are four combinations, and four comparisons are required. As a result of the comparison, the block is changed to M3,
Only when the block is replaced with M2, the corresponding combination exists on the conversion table 64. In this case, it is decoded into the number "9" corresponding to the combination. Third
The digit has a border bar E2 in the block and a border bar E1 in the block. The boundary bar E1 is the reference bar M1
And M2 are adjacent, the block is replaced by M1 or M2 to see if the table has the appropriate combination. In this case, when the block is replaced by M2 and at the same time the block is replaced by M2, it corresponds to the number "2", and when the block is replaced by M3 and the block is replaced by M1 the number "2".
9 ". In this way, when the boundary bar is replaced with the adjacent reference bar and the table comparison is performed, when there are two or more corresponding numbers, it is not known which one should be decoded, so it is treated as unconvertible. However, it is rare that two boundary bars are present in one digit, and the probability that the comparison result will be applied to two or more codes is further reduced, so that the reading accuracy is significantly improved. .

Second Embodiment Next, a second embodiment in which the decoding process is performed by using fuzzy inference will be described. For convenience of explanation, the type of reference bar, the type of parity code to be converted, and other conditions are the same as in the first embodiment. Further, in the fuzzy inference of this embodiment, a fuzzy controller MB94140 manufactured by Fujitsu Limited is used as a fuzzy chip.

FIG. 10 is a block diagram showing the structure of the decoding means 7 to which fuzzy inference is applied. The decoding means 7 includes a fuzzy inference processing unit 71 and a membership function storage unit 7.
2. A fuzzy rule storage unit 73 and a decode value determination unit 74 that determines and stores a decode value from the result obtained by inference. The decoding means decodes the corrected data in units of each digit. Figure 11
Shows the membership function of the input. Although only one membership function is shown in FIG. 11, four similar membership functions are set for each block. Figure 11
The horizontal axis of represents the input value, and the vertical axis represents the degree (weight) of belonging to the input variable. The input value is the corrected bar width count value. M1 corresponding to each reference bar as an input variable
M4 is selected. The input set value is determined according to the sizes of the reference bars M1 to M4. In decoding by fuzzy inference, the width of the standard bar that is normally assumed is not used as the input setting value, but the degree (weight) that belongs to each input variable M1 to M4 according to the possibility of reading error of the bar width. Is set so that adjacent input variables overlap with each other (for example, in M1 and M2, the count value 13
~ 17 overlap each other). As a result, an output corresponding to the weight of each input variable is output. 12
Shows the membership function of the output. The horizontal axis of FIG. 12 represents the output value and the vertical axis represents the weights of the output variables C0 to C9. The membership function for input and the membership function for output are stored in the membership function storage unit 72.
Memorized in. The fuzzy rule storage unit 73 stores a fuzzy rule (hereinafter referred to as a rule) for inferring an output value (decoded value) from a combination of the membership functions M1 to M4 of the input. FIG. 13 shows an example of fuzzy rules. For example, from rule 1 in FIG. 13, it is understood that when the blocks 1 to 3 are M3M2M1M1, the output is C0. This rule is defined according to the bar code standard and type. FIG. 13 shows JAN
This is the rule for the odd parity code of the code.

A more detailed specific example of the inference processing will be described with reference to FIGS. 6 and 13 to 18. The numbers with arrows shown above the membership functions in FIGS. 14 to 15 represent the corrected count value of each block used as the input value of the fuzzy inference. When the correction data is input to the fuzzy inference processing unit 71, the processing unit 71
Calls the corresponding membership function and fuzzy rule from the membership function storage unit 72 and the fuzzy rule storage unit 73, and performs inference processing for each rule.
First, the case where the input value (30, 10, 10, 20) for each block of the first digit in FIG. 6 is input will be described. FIG. 14 shows the relationship between the input and the membership function in this case. As can be seen from FIG. 14, the input variable M
There is no block that overlaps two areas 1 to M4, and the weight of all input variable applicability is 1. That is, in this case, each of the first digits corresponds to a combination of M3M1M1M2. This combination is inferred based on the rules shown in FIG. In this example, the input value applies only to rule 10 and does not apply to other rules. When the rule is not met, the output weight is 0. The output state of the first digit is shown at the top of FIG. As is apparent from FIG. 18, in this case, the first digit corresponds to "C9".
Further, since the weights of the input variable applicability of each block are all 1, the output weight is 1.0. Next, the case where the second digit count values (25, 9, 12, 24) in FIG. 6 are input will be described. In this case as well, inference processing is performed in the same manner as the first digit, but as can be seen from FIG. 15, in this case, the block and block input values span the regions of M2 and M3, respectively, and the weights of M2 and M3 are different. ing. In this case, four combinations with different weights, M2M1M1M2, M2M1M1M3, M3M1M
There are 1M2, M3M1M1M3. However, in this case, among the four combinations, only M3M1M1M2 corresponds to the rule 10, and the other combinations do not correspond to any of the rules. Therefore, this digit also corresponds to "C9".
This output state is shown in the second row of FIG. In this case, since the weight of the input variables M2 and M3 with the block is less than 1.0, the weight of the output variable is less than 1.0 (0.5 in FIG. 18).
3). Next, the case where the count value (24, 10, 15, 21) of the third digit in FIG. 6 is input will be described. In this case, as shown in FIG. 16, the block spans M2 and M3, and the block spans M1 and M2. In this case, the two rules are applicable, so a detailed description will be given with reference to FIG. The four vertical solid lines in FIG. 17 represent the input values (24, 10, 1) of the third digit.
5, 21), and the trapezoid indicated by the solid line indicates the input variable required by the rule. Therefore, if all four vertical straight lines intersect with the solid trapezoid, there is rule conformity. As can be seen from the above description and FIG. 17, this third digit corresponds to two rules, rule 3 and rule 10, and the output variable “C2” with different weights.
And "C9" are obtained. This output situation is shown in the bottom row of FIG. Weights (C2: 0.
53, C9: 0.27) varies depending on the weight of the input variable. The weight of the input variable is represented by the intersection of the vertical solid line and the trapezoid. The weight is 1 when the trapezoid intersects at the upper side, and the weight is less than 1 when the hypotenuse intersects.

The decode value (numerical value) of the output value output from the fuzzy inference processing unit is fixed by the decode value fixing unit 74. When the number of output variables of each digit input from the fuzzy inference processing unit 71 is one, the decoded value determination unit 74 determines the corresponding number from the number conversion table shown in FIG. Output variable is 2 like the 3rd digit
When there are two or more, the decoding may be performed with the higher weight or the decoding may not be performed. However, since the probability of erroneous decoding is high when the weight is used for determination, it is safe to process as undecodable unless the difference in weight is considerably large. In this embodiment, the first and second digits are fixed as the numeral "9", but the third digit is not fixed. In the above description, J
Although the decoding process using fuzzy inference for the left digit part of the AN code has been described, the decoding process using fuzzy inference is similarly performed for the start bar, the center bar, the stop bar and the right digit part. Be done. When the decoding process of the entire bar code is completed as described above, the decoded data is sent to the terminal device or the like via the output unit (not shown).

In the first and second embodiments, when the decoding process is performed, all the count values of the four blocks within one digit are used. However, as is clear from the table of FIG. 9 or the fuzzy rule of FIG. 13, even if any one of the blocks is deleted and only three arbitrary blocks are combined, the same combination or rule is not obtained. Therefore, it is possible to perform the decoding process according to the table or the rule using the count values of the three blocks in one digit. When a specific character can be specified without using all the elements that compose a digit in this way, the conversion table or rule may be defined by only the minimum necessary elements corresponding to the bar code. By performing the decoding process with the table or rule using the minimum number of elements, the comparison time or the inference time is shortened and the decoding process is speeded up.

[0021]

As described above, in the bar code reader according to the present invention, the bar code is corrected to a standard size in digit units, and the bar type is determined from the bar combination of all digits. Therefore, even if the barcode printing is slightly off the standard, dirty, or even if it is printed or affixed to a curved product such as fresh food wrapped in wrap and the barcode is distorted, it can be read. The range was expanded and the reading rate was greatly improved. For example, in the prior art, the radius of curvature is 30
Although it was possible to read only up to the state of distortion by mm, in the present invention, it became possible to read up to the state of distortion of 13 mm in radius of curvature. In addition, the bar width for one module of the standard bar code is 0.33 mm.
Then, in the conventional reader, the bar width is 0.17-0.
Although it was possible to read only up to 47 mm, according to the present invention, it became possible to read up to 0.09 to 0.55 mm, and the readable range was greatly expanded. Further, in the case of a hand-held bar code reader, conventionally, the position at which the reading section is separated by about 20 mm from the surface of the bar code is the limit position of reading, but in the present invention, it is possible to read even at about 30 mm.

[Brief description of drawings]

FIG. 1 is a block diagram showing an outline of the entire configuration of a barcode reading apparatus of the present invention.

FIG. 2 shows a digitally converted waveform of a barcode and a measured count value.

FIG. 3 shows a state in which measured count values are stored.

FIG. 4 is a flowchart showing a processing procedure of a correction processing operation of a measured count value.

FIG. 5 shows a specific example of a correction process when correcting to a reference digit length L = 70.

FIG. 6 shows a state in which corrected data is stored.

FIG. 7 shows the domain of each reference bar and boundary bar.

FIG. 8 shows a table showing bar combinations for each digit.

FIG. 9 is a conversion table showing a correspondence relationship between bar combinations and numbers.

FIG. 10 shows a configuration of a decoding means 7 to which fuzzy inference is applied.

FIG. 11 shows a block input membership function.

FIG. 12 shows an output membership function.

FIG. 13 is a fuzzy rule showing a relationship between a combination for each block and an output variable.

FIG. 14 shows a state of a membership function for inputting first digit correction data.

FIG. 15 shows a state of a membership function of inputting second digit correction data.

FIG. 16 shows a state of a membership function for inputting third digit correction data.

FIG. 17 shows a state of an input / output membership function of each fuzzy rule of the third digit.

FIG. 18 shows the states of the output membership function of the first to third digits.

FIG. 19 is a conversion table showing a relationship between output variables C0 to C9 and numbers of JAN code.

[Explanation of symbols]

 3 signal processing means 5 signal correcting means 6 decoding means 7 decoding means

Claims (5)

[Claims] 1. A signal generating means for outputting a bar code read signal according to the width of a bar and a space of a bar code, and a width of each bar and space from the bar code read signal.
And a measuring means for measuring and storing the length of each digit composed of a plurality of bars and spaces, and the read signal according to the ratio between the previously stored regular digit length and the measured digit length. Of the bar code and the width of the space are expanded or reduced in digit units to correct the read signal, and a decode unit that decodes the read signal corrected by the correction unit in digit units is read. apparatus. 2. A signal generating means for outputting a bar code reading signal according to the width of the bar and the space of the bar code; a measuring means for measuring and storing the width of each bar and the space from the bar code reading signal; The read signal is obtained by dividing the type of each bar and space according to the width of the bar and space measured by the measuring means, and comparing the combination of each divided bar and space with a predetermined table. Decoding means for decoding, wherein the decoding means further comprises: when the measured width of the bar or space is close to the boundary of the compartment,
Boundary data processing means for attempting decoding processing when the bar or space is divided for each of the two sections adjacent to the boundary, and processing is considered as correct decoding only when only one decode value is determined for both sections. Bar code reader including. 3. The decoding means according to claim 2, comprising: the measuring means according to claim 1; and the correcting means according to claim 1, wherein the decoding means decodes with the read signal corrected by the correcting means. The bar code reader according to claim 2, wherein the bar code reader is provided. 4. A signal generating means for outputting a bar code reading signal according to the width of the bar and the space of the bar code, a measuring means for measuring and storing the width of each bar and the space from the bar code reading signal, A bar code reading device comprising: a decoding means for decoding the read signal by fuzzy inference using the widths of the bars and spaces from the measuring means as input information and the combination of types of bars and spaces as a rule. 5. The decoding means according to claim 4, further comprising a measuring means according to claim 1 and a correcting means according to claim 1, wherein the decoding means decodes with the read signal corrected by the correcting means. The bar code reader according to claim 4, wherein the bar code reader is provided.