Control-Free Low-Level Image Segmentation: Theory, Architecture, and Experimentation

4 stars based on 40 reviews

To receive news and publication updates for Journal of Engineering, enter your email address in the box below. Box 7, Ranchi, JharkhandIndia 3 A. This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use, distribution, and reproduction in any medium, provided the original work is 4341 feature binarization cited. A number of techniques have been proposed earlier for feature extraction using image binarization.

Efficiency of the techniques was dependent on proper threshold 4341 feature binarization for the binarization method.

In this paper, a new feature extraction technique using image binarization has been proposed. The technique has binarized the significant bit planes of an image by selecting local thresholds. The proposed algorithm has been tested on a public dataset and has been compared with existing widely used techniques using binarization for extraction of features.

It has been inferred that the proposed method has outclassed all the existing techniques and has shown consistent classification performance. Incessant expansion of image datasets in terms of dimension and complexity has escalated the requirement to design techniques for 4341 feature binarization feature extraction.

Selection of image 4341 feature binarization has been the basis for content based image classification as reviewed by Andreopoulos and Tsotsos in [ 1 ].

In this work, a new feature extraction technique applying binarization on bit planes 4341 feature binarization local threshold technique has been proposed. A digital image can be separated into bit planes to understand the importance of each bit in the image as shown by Thepade et al. The process was followed by binarization of significant bit planes for feature vector extraction.

Binarization process calculated the threshold value to differentiate the object of interest from its background. The novel method has been compared quantitatively with the techniques proposed by Thepade et al. Mean square error MSE method was followed for classification performance evaluation of the proposed technique with respect to the existing techniques for feature vector extraction. Various methods have been used for feature extraction that has implemented image binarization as a tool to denote the object of interest and its background, respectively.

Threshold selection has been essential to facilitate binarization of image to differentiate the object from its background. Process of threshold selection has 4341 feature binarization categorized into three different techniques, namely, mean threshold selection, local threshold selection, and global threshold selection. Existing methods of feature extraction from images using selection of mean threshold were adopted by Thepade et al.

The first method of feature extraction using even and odd images [ 2 ] has generated two different 4341 feature binarization of images by adding and subtracting the original image and its flipped version, respectively, for each variety as shown in where denotes the original image and denotes its flipped version.

Binarization of even image and odd image were done by mean threshold technique as shown in 2 and the feature vectors of size 12 were generated. In the second case of feature extraction, noteworthy information was extracted from the images by comparing the values of significant bits in each pixel and the extracted values were binarized following the process of mean threshold selection for binarization [ 2 ]. Equation 2 shows the process of mean threshold selection.

Consider where is redgreenand blue for each color component. Feature vectors of dimension 12 were generated by taking the mean of significant values higher than the threshold and 4341 feature binarization than the threshold, respectively. The third method for feature extraction with multilevel block truncation coding [ 3 ] was an iterative process.

The initial stage started with binarization using mean thresholding as in 2 and two feature vectors for 4341 feature binarization color component were generated from the mean of gray values higher than the threshold and lower than the threshold, respectively. Classification performance was tested with the generated feature vectors at the first level of feature vector extraction. The next step involved binarization of the gray values higher than the mean threshold and lower than the mean threshold at the first 4341 feature binarization.

This has produced four feature vectors for each color component and the classification performance was evaluated with the generated feature 4341 feature binarization at second level. Now, the classification results for the first level of feature extraction and second level of feature extraction were compared to find out improvement in performance.

If the result for second level was found to be higher than the first level, 4341 feature binarization the entire feature extraction process was repeated for further levels of feature extraction. The process was continued until the classification performance deteriorated. Multilevel block truncation coding method of feature extraction has shown its best performance at level 3 and the feature vector size at the mentioned level was The drawback of mean threshold method was to only determine a midpoint and has not been effective to differentiate the spread of data among the datasets.

The methods involved sliding of a window over the image for calculation of threshold value pixel-wise. Binarization process was performed locally by comparing each pixel value to the corresponding threshold. A pixel value greater than the corresponding threshold was represented by 1 and a lower value than 4341 feature binarization was denoted by zero.

The threshold for each color component was computed as the mean of the highest and the lowest gray value within the local window specified.

The contrast was expressed as the distinction between the greatest and the least gray values. Binarization was done by comparing 4341 feature binarization contrast value to the contrast threshold.

4341 feature binarization method categorized the pixels in an image 4341 feature binarization background and foreground pixels by calculating the optimal threshold. The drawback for the techniques was the hefty feature vector size generated by each. The dimension of feature vectors was dependent on the size of the image. Thus, for an image of sizethe generated feature vector size was.

Elalami [ 11 ] has extracted color and texture based features by 3D color histogram and Gabor filters and addressed the problem for large feature vector size with genetic algorithm. Hiremath and Pujari [ 12 ] have divided the image into nonoverlapping blocks. Local descriptors of color and texture were calculated from the 4341 feature binarization moments and moments on Gabor filter responses of these blocks were considered to generate local descriptors of color and texture.

Gradient vector flow fields were used to define the edge images 4341 feature binarization shape descriptors followed by the use of invariant moments. Jalab [ 14 ] has combined color layout descriptor and Gabor texture descriptor as a fusion approach for better identification of images.

Shen and Wu [ 15 ] have extracted feature vectors from images by exploiting color, texture, and spatial structure descriptors. Rahimi and Moghaddam [ 17 4341 feature binarization have explored the intraclass and interclass features for effective extraction of 4341 feature binarization signatures. The proposed technique was compared to seven other state-of-the-art feature extraction techniques for assessment of classification performance. The quantitative evaluation and statistical analyses have revealed the efficiency of the proposed method.

The proposed method in this work can be subdivided into two different levels, namely, Bit Plane Slicing and threshold selection as given below. Individual bits needed to represent a number in binary have been considered as the bit planes. The first level started with segmentation of the images 4341 feature binarization various bit planes starting from the least significant bit LSB to the most significant bit MSB.

The intensity value of each pixel was denoted by an 8-bit binary vector and the value was either 0 or 1 as shown in Figure 1. Thus, each bit 4341 feature binarization was represented by a binary matrix which was further used to generate image slices for respective bit planes as in Figure 2. Significant bit planes carrying rich information were identified and considered for feature extraction by binarization as in 3 and insignificant bit planes were discarded.

Consider where is, andrespectively, for each color component. The test dataset contains different scenes from diverse categories having varying illumination across the scene as observed in Figure 7. Uneven luminance distribution caused the image to be brighter or darker at 4341 feature binarization which has adversely affected the recognition of the object of interest.

Thus, choosing a single global or mean threshold for the entire scenes of diverse categories was 4341 feature binarization an effective alternative. Instead, selection of local threshold technique can handle the problem of uneven illumination by selecting threshold locally. Pixel-wise threshold was calculated by the method for each color component of the selected bit planes by sliding a rectangular window over the component.

The threshold was determined based on the local mean and standard deviation. The window size was considered as. The threshold was given as. Here, has 4341 feature binarization a constant having value between 0 and 1 and was considered to be 0.

The value of and the size of sliding window determined the quality of binarization. Figure 3 a has shown the original image from which bit plane 5 has been extracted in Figure 3 b. Figure 3 c 4341 feature binarization shown the effect of binarization on the original image using mean threshold.

The image of bit plane 5 has been binarized with mean threshold in Figure 3 d. Begin 1 Input an image with three different color components 4341 feature binarization,andrespectively, of size each. The proposed technique was experimented with a subset of Corel stock photo database known as Wang dataset used 4341 feature binarization Li and Wang in [ 18 ].

It has been considered as a 4341 feature binarization used public dataset. The experimental process has not compromised with the size and quality of the 4341 feature binarization. The dataset comprised 9 categories with images in each category. Figure 7 shows a sample of the original database. The classification performances were evaluated with fold cross-validation scheme for the proposed and existing feature vector extraction techniques.

The process has been called -fold cross-validation as 4341 feature binarization by Sridhar in [ 19 ]. The value of is an integer and was considered 4341 feature binarization be 10 in this 4341 feature binarization. The entire dataset was divided into 10 subsets. The method was repeated for 10 trials and the performance of the classifiers was evaluated by combining the 10 results thus obtained after evaluating 4341 feature binarization 10 folds. The proposed technique for feature extraction has been evaluated by mean square error MSE method as in The method considered the similarity measure between two instances as described 4341 feature binarization Xu et al.

The nearest neighbor in the instance space was located for classification and then the unknown instance was designated with the same class of the identified nearest neighbor. Consider where and are the two images used for comparison using the MSE method. Primarily, two different evaluation metrics, namely, misclassification rate MR and score, were considered to compare the proposed feature extraction technique with respect to the existing 4341 feature binarization.

The error rate of the classifier indicates the proportion of instances that have been wrongly classified as in where true positive TP is number of instances classified correctly, true negative TN is number of negative results created for negative instances, false positive FP is number of erroneous results as positive results for negative instances, and false negative FN is number of erroneous results as negative results for positive instances.

Precision and Recall TP rate can be combined to produce a metric known as score as in It has been considered as the harmonic mean of Precision and Recall. Higher value of score indicates better classification results.

Consider where Precision is the probability that an object is classified correctly as per the actual value and Recall is the probability of a classifier that it will 4341 feature binarization true positive result. The comparison shown in Tables 1 and 2 has been graphically represented in Figures 8 and 9.

It has been observed that the images of category named dinosaur 4341 feature binarization got minimum misclassification rate and the highest score.

Y las opciones de 60 segundos

  • Trading systems and money management

    Us friendly binary options brokers which accept traders from the united states

  • What are binary options trading discussions

    Best free forex signals site

What is the best binary options indicator

  • A four-month european call option on a dividend-paying stock is currently

    Mustafa forex ringgit

  • Canadian dollar forex chart

    Trader forex sukses dari indonesia

  • Heber binary plants imperial valley california usac

    Mortgage broker trade magazine canadian

High frequency trader salary glassdoor

13 comments Senarai broker forex malaysia

10 375 in binary option strategies

Postal nine minutes mechanism The invention relates to postal matter nine minutes apparatus using that, more specifically, which makes it possible to simplify the system by identifying the mail item by using the feature information of the postal image and reduce the maintenance costs, the postal image feature on the mail item nine minutes device.

Looking at the domestic mail delivery system, separate from the delivery shipped distinction is made in the postal mail jipjungguk, where nine minutes mailing process by using a ZIP code as described in the mail being processed by an automated mail classification. Deliverymen Circuit in accordance with a predetermined path within the mail delivery area when delivery of mail and to deliver the mail, and this path is called a baedalsun.

Deliverymen called nine minutes fairway this process to sort mail in order to baedalsun of mail prior to delivery of postal water. On the other hand, it is referred to as fairway sorter machine to sort the mail item in the sequence shown in baedalsun. Unlike existing group fairway nine minutes of mail groups and nine minutes to complete the double over nine minutes fairway classified through the mail operation, array of procedures is based on the principle of radix sort algorithm.

Mails that were made in the process of nine minutes ZIP jipjungguk is contrary inde process classified into different bundles defined mail, mail fairway nine minutes the process is different in that it is the process of sorting the mail in a series of predetermined sequence. Another difference of nine minutes tile-based mail fairway nine minutes distinction mailing group that is primarily that, while being separated by a zip code, a sort can be precisely group fairway nine minutes to read address information, postal address.

Thus, due to features such as the automation of mail order nine minutes per fairway mail delivery address in the nation, and needs to be assigned a unique number, a unique number such as baedaljeom code or address code.

Therefore, it was an effective alternative to the current system, to adopt a barcode if you want to achieve the fairway nine minutes working group to leverage traditional mail segment. This is the whole address reading process by printing a bar code on the mail with the address read during the first mailing nine minutes operation, and the second mailing nine minutes obtaining the address information is read prior to reading a printed barcode during operation is limited to one to.

In this case, call the barcode information is printed on the mailing that mail ID. Postal nine minutes apparatus using a conventional bar code technology, postal nine minutes mechanism, also see Fig. Copper jakye cases to distinguish fairway mail by the conventional bar code reading and printing techniques are as follows. After the first nine minutes of the second mail mailing nine minutes step is completed, the operator then collect the mail from the mail load compartment in a specific order, and re-introduced a mechanism to collect mail on postal nine minutes to start the second stage mailings nine minutes.

Two if the second postal matter nine minutes mail is input in the step, the control member postal nine minutes obtain a mail serial number from the bar code reading section of the mechanical portion and the first mail item nine minutes memory using a mail serial number, an address code that was saved at step obtained from the apparatus, and determines a destination address corresponding to the load compartment, depending on fairway code plan, and the subsequent operation is the postal matter nine minutes and proceeds in the same manner as in the first nine minutes mail steps: In summary, the process fairway nine minutes using a conventional bar code as described above, first, print nine minutes in the mail process mail ID with the address read by the bar code, and stores the read address results for mail ID in the storage device.

Second, since the postal matters are to nine minutes to process without reading the address reads the mail ID barcode obtains the address of the read result from the storage device to the mail ID and separate the mail address according to the read result. However, the conventional method for identifying the mail item by using the bar code on the mail nine minutes apparatus as described above, there is the need for a separate bar code printing and reading device for the mailing nine minutes device, bar code printing apparatus as a black and white bar code printing apparatus or fluorescent use barcode printing device.

Black and white bar code printing apparatus has a problem in that the read out rate is significantly different depending on the material state and the printing state of the postal matter nameuro difficult to apply the fairway nine minutes problem of mail. On the other hand, the fluorescent bar code printing device tricky management of bar code printers methods that require a professional engineer with separate qualifications for the treatment and management of chemicals, such as fluorescent ink, keep on consumables such as fluorescent ink costs since there is a problem that even a high weighting for economic burden on the operation of the mail devices nine minutes.

The invention with be proposed by using the feature information of the postal image by identifying the mail, which will achieve such simplified systems, maintenance costs, fairway nine minutes improve the success rate, the postal image feature in order to solve the above problems to provide a mail nine minutes device they are an object.

It may be understood by the following description of Other objects and advantages of the present invention will be appreciated more clearly by the embodiment of the present invention. Also, the objects and advantages of the invention will be readily appreciated that this can be realized by the means as claimed and combinations thereof. And the postal image identifying means for transmitting the mail by extracting the image feature of a postal image received from the image processing means, and giving the video-ID number with respect to the mail image by using an image characteristic of the postal image as the control means It includes.

The aforementioned objects, features and advantages will become apparent from the following description in conjunction with the accompanying drawings, a self-technical features of the present invention one of ordinary skill in the art thus can be easily There will be. Further, the detailed description of the known art related to the invention In the following description of the present invention will be omitted and a detailed description on the case that are determined to unnecessarily obscure the subject matter of the present invention.

It will be described in the following, preferred embodiments of the present invention with reference to the accompanying drawings in detail. Figure 2 is one embodiment of the configuration of mail nine minutes apparatus using a mail image feature in accordance with the present invention.

Mail nine minutes device according to the invention, as illustrated in Figure 2, the mail dividing means as a postal nine minutes mechanism part , a control unit as control means , the postal image processing as a mail image processing means , address read It comprise the means as a read address unit , and the postal image as the postal image identifying means identifying part Then, the mail nine minutes mechanism Fig.

The present invention as described above, unlike in the fairway nine minutes the process of automatically sorting the mail according to the delivery order of the mail delivery person reads the address of the recipient according to the mail, the prior art requiring the bar code reading and printing characteristics of the postal image using the information by identifying the mail relates to the postal nine minutes apparatus and method having an effect to simplify and reduce the maintenance costs for the structure of the mail unit nine minutes.

Referring to the key role portion each configuration, the postal nine minutes mechanism while conveying the mail item through the conveying belt , moving at a high speed by determining the path of the mail item with the diverter , the mechanism to separate the mail item, the control unit controls the operation of the W-mail to determine the loading compartment mail arrives along the fairway plan of mail nine minutes mechanism part , the postal image processing unit transmitting a postal image on the mail nine minutes mechanism after the image processing such as image binarization process to obtain a postal image from the line scan cameras in the postal image obtaining unit of the address reading section , and the address read out unit reads the recipient address from the postal image and outputs an address code corresponding to the address reading.

In terms of comparison with the following, prior art see Fig. According to the present invention does not print or read the bar code, but instead stores the feature information of the mail item from the first postal matter nine minutes steps in the memory device, and by both comparing the feature information of the first in the mail nine minutes Step mail address code information for the mail item to be obtained.

In the present invention requires a postal image identification unit in place of a bar code reader or a bar code printing press, an operation example in the case of separating the fairway mail using this are: When the first postal matter nine minutes mail is input in the step, and controller is assigned to a mail serial number, and transmits a mail serial number on the postal image processor , the postal image processing unit is a postal image input from the line scan camera a binarized and then, mailing serial number and the binarized mail image the address reading section , the postal image and respectively sent to the identifying part , the address reading section and then reads out the payment address for mailing the address transmitted to the control unit with the code and the mail serial number, and control unit is the destination loading compartment information in determined by the fairway planning a destination loading compartment and postal nine minutes mechanism corresponding to the address code of the mail item transfer, postal nine minutes mechanism transfers the mail to the destination load compartment.

Figure 3 is one embodiment of a detailed configuration of the mailing nine minutes mechanism of Figure 2 in accordance with the present invention, Figure 4 is an embodiment schematic of mail branch and mail loading portion in the postal nine minutes mechanism of Figure 2 in accordance with the present invention.

Postal nine minutes mechanism as mail dividing means is, according to the, to the mail to be nine minutes sequentially input to the mail nine minutes mechanism 'mail input unit , the mail transfer belt as shown in Figure 3 obtaining a postal image from being conveyed mail, the postal image acquiring unit , 'mail branch, which by changing the conveying path of the mail item to be stacked mail in any mail stacking space , a plurality of postal matter loading It is configured to include along a transfer belt, the mail loading part , and the rotation consisting of a space for transporting the mail to a predetermined conveyance path, mail transferring portion Referring to the role of each configuration unit of the mail nine minutes mechanism part , the mail item input unit , and to put into a mail nine minutes mechanism to give a piece of a predetermined interval, or a period of time the number of sheets of mail in sequence, the postal image obtaining unit has various optical for positioning of the postal matter is transmitted to mail the image information on the postal image processor then acquires the mail image by using a line scan camera to change the postal image information into digital information, and mail transferring the sensor such as various rollers for moving the , and the mail feed belts is provided.

Mail branch , the mail loading section , when the mail item transport unit is described with an example of the interaction under the control of the control unit to Fig. Controller as control means, as shown in Figure 5, the mail location tracking unit , a mail serial number generation unit , a mail serial number transmission section , the address reading result receiver , including the mail loading compartment determination section determination mail stacked position , a mail branch signal generator , the ID number candidate generator , the identification number candidate transmission unit , the identification number receiving by made.

Operation of the control unit , the first operation is different from the first mail item when nine minutes during first mail nine minutes and a second mail item separated when the second mail matter.

First, the first will be described an operation of the second postal matter nine minutes step claim 1 when mail sensitive controller in, as follows. Mail tracking section tracks the position at which the transfer along the two mail path receives the signal from various kinds of optical sensors are attached to various locations on the mail item transport unit of the mail nine minutes mechanism part in real time.

Mailing serial number generation unit generates a serial number to refer to each mail item being transferred along a transfer path by mail the image processor , the address reading section , the postal image identification unit , mailing serial number transmission section transmits the mail created by the mail serial number serial number generating unit at the right hand image processor Address reading result receiving unit addresses the read and receive the results of the sub-address read at , the mail loading space determining unit addresses the read result and fairway nine minutes plan information mail is mail mounting portion , using the of determining whether or be separated by any loading space, and to configure a mail branch signal generator is the mail branch of the mail nine minutes mechanism according to the result determined in the mail loading compartment determiner It generates a control signal for operating a series of the diverter , At this time, the result determined in the mail loading space determining unit is stored in the memory for reference in the second mail item identification number candidate generator during classification.

The reason for giving a mail serial number from control unit to all the mail being conveyed along a mail item transport path in a mail nine minutes mechanism part in real time to the mail location tracking unit of the controller that the mail is transferred tracking while the mail image processing unit , the address reading section , the postal image identification unit in, regardless of the mail item position on the conveying path of the postal matter because it does not have position information of a mail item in real time as set, and the first regardless of whether the first nine minutes mailing step or a second step after nine minutes of mail, because it requires a means for referring to the mail item.

On the other hand, the two will be described an operation of the control unit of the mail item in the second stage nine minutes second si mail classification , as follows. Mail location tracking unit and the mail serial number generator in the first mail item sensitive and the same operation as that in step, the identification number candidate generator is loading the mail item from the first postal matter nine minutes Step mail compartment using a loading results by estimating the second order of the mail in the mail inputted nine minutes step generates the image identification number of the candidate by the predetermined number.

Identification number candidate transmission unit transmits the image identification number candidate generated by the ID number candidate generator , the identification number candidate receiving portion of the postal image identifying unit , the identification number receiving receives the identification number determined by the postal image identification unit , the mail loading space determining unit searches the address reading result obtained in the first mail item nine minutes steps from the video-ID number to address the read result as a fairway nine minutes plan information used to determine whether mail item is to be separated by any loading compartment on the mail loading part , and the mail branch signal generator operates in the same manner as in the first nine minutes mail steps: As postal image processing unit as a postal image processing means is shown in Figure 6, the first postal image receiving unit , a serial number receiving unit , the postal image sequence number assigning unit , the postal image transmission unit , it comprises a first postal image binarization section , the postal image compression unit Referring to the basic operation of the postal image processing unit , a first postal image receiver ZIP nine minutes receive incoming mail images from the postal image obtaining unit of the mechanical portion , a serial number receiving unit receives receiving a mail serial number from a mail serial number transmission section of the controller , the postal image serial number assignment unit in consideration of the received sequence and the reception time of the postal image and a mail serial number, the postal image and the mail item sikimyeo the serial number one-to-one basis, the postal image transmission unit transmits a mail serial number and the mail address to the image reading section , the postal image identification unit , respectively.

At this time, it is possible to a data compression prior to sending the mail image in order to shorten the time of transmitting a postal image as the postal image processor , the address reading section or the postal image identification unit from, for this purpose postal image processing unit may include a first postal image binarization section , the postal image compression unit to the secondary.

Here, the first postal image binarization unit is a data length required for binary ingestion mail image by determining the pixel value of the postal image into black or white, and the postal image compression unit to transmit the mail image information to reduce, change to the compressed mail to the binarized image information expressing form.

Accordingly, the address reading section is additionally sent to the image may include a restoration unit , the postal image identifying unit is the ability to restore the mail image in the image feature extraction unit or the postal image the may include the ability to binary. Address reading means as an address reading section as shown in Figure 7, the second postal image receiving unit , the postal image restoring unit , a second postal image binarization section , an address area determination section , comprises a character string extraction unit , a character extracting unit , a character recognition section , address analysis section , a read result transmitting unit Referring to the operation of the address reading section , a second postal image receiving unit receives receiving the mail images from the postal image transmitting section of the postal image processor , the received mail video compression and binarization the case image "", "when the binarization is compressed in , the postal image and restoration unit restores through decompressed , and brightness for the postal image " If you do not do not go through the "binarization , the second postal image binarization is binarized through the unit After the process to is a conventional address, the read process, the address area determination section , a character string extraction unit , a character extracting unit , a character recognition section , address analysis section the address code is output through.

That is, the address area extracting unit includes an image of a portion in which the recipient address base from the postal image payment address image extracted, and the character string extraction unit extracts a character string image from the extracted recipient address image to If, character extraction unit extracts each character image from the character string image.

Then, the character recognition section reads out the individual character images by looking output a character code and a confidence value corresponding to a Hangul or English, number and address analysis section is a combination of a series of character codes and confidence values try to find the most suitable address code output from the address dictionary.

Reading result transmitting section transmits the address code output from the serial number and the mail address analysis section received with the reception when receiving mail from the mail video image processor , the control unit Figure 8 is one embodiment of the detailed construction, zip receive input mail images from the video processor , the postal image identification unit to identify the postal image as in Figure 2 of the postal image identification unit according to the invention a indicates the function.

Postal image identifying means as the postal image identification unit , as shown in Fig. Operation of the postal image identification unit operates differently when the second mail item when separated from the first postal matter separated.

First, the first will be described an operation of the second postal matter nine minutes step claim when one postal matter sensitive ZIP image identification unit in the third postal image receiving the postal image transmission section of the postal image processor , mail serial number, the postal image mail receive receives the postal image on which the serial number has been assigned, the image feature extraction unit extracts the characteristic information that can identify the postal image from the postal image, image feature registration unit from stores the serial number and the image feature information of the image on the video mail feature store.

At this time, the mail reason to limit the search to the image feature information and limited to image identification number candidate in an image identification unit geotinde to shorten the search time required to select the image feature information, search of the image feature information If the limit is too much to be affected in the double-feed problems caused by mechanical failure or loss of mail due to the negligence of the operator.

Thus, the postal nine minutes mechanism , the mail sorting operation speed, the mail to be image identification number candidate generated by the frequency of occurrence of double-feed, consider the loss costs for lost problem of mail by the identification number candidate generator of the appropriate together crystal is preferable to, when the postal matter sorting operation speed of the postal nine minutes device for example 30, barrel and out a double feed occurrence frequency is 0.

At this time, when using the postal image reading characteristics, such as PO, and AR, the implement some functions of the address reading section , the image feature extraction unit , generate the address read process from the address reading section it is necessary to transmit the result mail to the image identification unit Image feature comparison unit compares the value of each feature extracted from the image feature extraction unit and the image feature information for each identification number candidates.

At this time, the comparison between the characteristics of the comparison with the vector type among the characteristic of the scalar type and outputs a distance value between the feature corresponding computed arithmetically by the Euclidean distance calculation, a numeric string distance comparison distance value between the following: Distance comparison between the different strings as compared to the AR-dimensional distance between the vector of the CP, such as may employ conventional dynamic matching algorithm in one embodiment.

Thus, each feature comparison value image feature distance value is converted to a scalar value, with the calculation method may select the following Equation 2 in a single embodiment. Here, W i indicates a weight for each image feature, and the distance, C i denotes a value for each image feature distance calculated above.

Identification number selecting unit , a total image feature distance D is the smallest calculated to output the identification number of the image feature, the smallest total image feature distance D if the value is beforehand is greater than the given threshold value T have the do not display the image identification number can dismiss any candidate identification number.

Therefore, the mail item loaded in the two spaces determiner is the mail is not output the image identification number in or not output the correct address code in the address reading section , the postal code identifying section ZIP nine minutes mechanism of it it is preferable to reject the mail to be separated by loading compartment provided on the mail item loading section Here, ceil x indicates the smallest integer not less than x, and, log x, y denotes the logarithm of x to y for the radix.

Referring to the characteristics of the radix sort algorithm using a fairway nine minutes mail process illustrated in FIG. At this time, the sort order of the mail item in the first delivery address for the convenience of calculation are intended to mark zero. On the other hand, the total number of baedaljeom in Figure 10 DP 25, when the total number NP of the normal loading compartment is assumed to be 5, the required postal nine minutes job number TP is a calculation example of "" in FIG.

That is, the input mail item indicated from 6 to 42 5 can be found for each separated by the first mail item Category 2 times the loading compartment in step, the second step in the mailing nine minutes loading compartment 4.

In the same process, look at the number system Pental, the first in the second mailing nine minutes step and separating the mail by the digit value of 15 for the number of mail sorting order, second mailing nine minutes steps based on the digit values of 10 5 as to separate mail is the same result.

Which loading of the ID number candidate generator , mail item identification number if the principle of determining the candidates described with reference to the explanatory diagram of Figure 10, the mail loading section with respect to all mail, separated from the first postal matter nine minutes steps by checking whether the loaded column in the mail location tracking unit , the first can be predicted the final state of the second mail item coded results , whereby after the first nine minutes mail is mail collected result is also calculated.

Therefore, assuming that an error such as double feeding mistakes or mail input of the worker does not occur after the first mailing distinct prediction mail order mail collection results and two mailings in the second mailing nine minutes steps in order to match, because the identification number candidate generator outputs a list of serial numbers for the front and rear of the N pieces of a predetermined input current postal mail.

In one example, the candidate number N as this is set to 5, when the predetermined current input mail item assumed to mail indicated by "" in Fig.

The method of the present invention as described above may be stored in a form that is implemented as a program in a computer-readable recording medium a CD-ROM, RAM, ROM, floppy disk, hard disk, optical magnetic disk, etc. Since the process can be easily implemented by those of ordinary skill in the art and will not be further described in detail. The present invention described in the above, the present invention are here to those of ordinary skill in the art can be various changes and modifications may be made without departing from the scope of the present invention, since the above-described embodiments and the accompanying It not limited by the drawings.

The present invention as described above, since without the use of a bar code printing machine identifies the mail by using the feature information of the postal image, since it requires a separate force for professional maintenance technician for the management of a bar code printing press, the holding of mail nine minutes device there is an effect that the maintenance easy.

In addition, the present invention is to simplify the mechanical construction of the mail nine minutes devices because they do not use bar code readers and bar code printing machine and and shorten the mail transport path and also the effect of allowing this reason to create a mail nine minutes device with a smaller size have.

In addition, the present invention, since it does not print the bar code, there is no consumption of the ink for printing the bar code, there is an effect that, due to the printing of the bar code on the mail item the outside does not occur a problem that postal matters are not marred. Year of fee payment: The present invention relates to a mail will nine minutes apparatus using a mail image feature.

The invention attempts to solve the technical challenges. The invention that is the object to provide an apparatus using a mail nine minutes, the postal image feature which makes it possible to simplify the system by identifying the mail item by using the feature information of the postal image and reduce maintenance costs.

Resolution of the subject matter of the invention,. And the postal image identifying means for transmitting the mail by extracting the image feature of a postal image received from the image processing means, and giving the video-ID number with respect to the mail image by using an image characteristic of the postal image as the control means including the.