Convolutional Neural Network (ConvNet) is one class of the powerful Deep Neural Network that can analyze and learn hierarchies of visual features. Originally, first neural network (Neocognitron) was introduced in the 80s. At that time, the neural network was not broadly used in both industry and academic field by cause of large-scale dataset shortage and low computational power. However, after a few decades later in 2012, Krizhevsky made a breakthrough on ILSVRC-12 visual recognition competition using Convolutional Neural Network. That breakthrough revived people interest in the neural network. The success of Convolutional Neural Network is achieved with two main factors. First of them is the emergence of advanced hardware (GPUs) for sufficient parallel computation. Second is the availability of large-scale datasets such as ImageNet (ILSVRC) dataset for training. Unfortunately, many new domains are bottlenecked by these factors. For most domains, it is difficult and requires lots of effort to gather large-scale dataset to train a ConvNet. Moreover, even if we have a large-scale dataset, training ConvNet from scratch is required expensive resource and time-consuming. These two obstacles can be solved by using transfer learning. Transfer learning is a method for transferring the knowledge from a source domain to new domain. There are two major Transfer learning cases. First one is ConvNet as fixed feature extractor, and the second one is Fine-tune the ConvNet on a new dataset. In the first case, using pre-trained ConvNet (such as on ImageNet) to compute feed-forward activations of the image into the ConvNet and extract activation features from specific layers. In the second case, replacing and retraining the ConvNet classifier on the new dataset, then fine-tune the weights of the pre-trained network with the backpropagation. In this paper, we focus on using multiple ConvNet layers as a fixed feature extractor only. However, applying features with high dimensional complexity that is directly extracted from multiple ConvNet layers is still a challenging problem. We observe that features extracted from multiple ConvNet layers address the different characteristics of the image which means better representation could be obtained by finding the optimal combination of multiple ConvNet layers. Based on that observation, we propose to employ multiple ConvNet layer representations for transfer learning instead of a single ConvNet layer representation. Overall, our primary pipeline has three steps. Firstly, images from target task are given as input to ConvNet, then that image will be feed-forwarded into pre-trained AlexNet, and the activation features from three fully connected convolutional layers are extracted. Secondly, activation features of three ConvNet layers are concatenated to obtain multiple ConvNet layers representation because it will gain more information about an image. When three fully connected layer features concatenated, the occurring image representation would have 9192 (4096+4096+1000) dimension features. However, features extracted from multiple ConvNet layers are redundant and noisy since they are extracted from the same ConvNet. Thus, a third step, we will use Principal Component Analysis (PCA) to select salient features before the training phase. When salient features are obtained, the classifier can classify image more accurately, and the performance of transfer learning can be improved. To evaluate proposed method, experiments are conducted in three standard datasets (Caltech-256, VOC07, and SUN397) to compare multiple ConvNet layer representations against single ConvNet layer representation by using PCA for feature selection and dimension reduction. Our experiments demonstrated the importance of feature selection for multiple ConvNet layer representation. Moreover, our proposed approach achieved 75.6% accuracy compared to 73.9% accuracy achieved by FC7 layer on the Caltech-256 dataset, 73.1% accuracy compared to 69.2% accuracy achieved by FC8 layer on the VOC07 dataset, 52.2% accuracy compared to 48.7% accuracy achieved by FC7 layer on the SUN397 dataset. We also showed that our proposed approach achieved superior performance, 2.8%, 2.1% and 3.1% accuracy improvement on Caltech-256, VOC07, and SUN397 dataset respectively compare to existing work.
The objective of this research is to test the feasibility of developing a statewide truck traffic forecasting methodology for Wisconsin by using Origin-Destination surveys, traffic counts, classification counts, and other data that are routinely collected by the Wisconsin Department of Transportation (WisDOT). Development of a feasible model will permit estimation of future truck traffic for every major link in the network. This will provide the basis for improved estimation of future pavement deterioration. Pavement damage rises exponentially as axle weight increases, and trucks are responsible for most of the traffic-induced damage to pavement. Consequently, forecasts of truck traffic are critical to pavement management systems. The pavement Management Decision Supporting System (PMDSS) prepared by WisDOT in May 1990 combines pavement inventory and performance data with a knowledge base consisting of rules for evaluation, problem identification and rehabilitation recommendation. Without a r.easonable truck traffic forecasting methodology, PMDSS is not able to project pavement performance trends in order to make assessment and recommendations in the future years. However, none of WisDOT's existing forecasting methodologies has been designed specifically for predicting truck movements on a statewide highway network. For this research, the Origin-Destination survey data avaiiable from WisDOT, including two stateline areas, one county, and five cities, are analyzed and the zone-to'||'&'||'not;zone truck trip tables are developed. The resulting Origin-Destination Trip Length Frequency (00 TLF) distributions by trip type are applied to the Gravity Model (GM) for comparison with comparable TLFs from the GM. The gravity model is calibrated to obtain friction factor curves for the three trip types, Internal-Internal (I-I), Internal-External (I-E), and External-External (E-E). ~oth "macro-scale" calibration and "micro-scale" calibration are performed. The comparison of the statewide GM TLF with the 00 TLF for the macro-scale calibration does not provide suitable results because the available 00 survey data do not represent an unbiased sample of statewide truck trips. For the "micro-scale" calibration, "partial" GM trip tables that correspond to the 00 survey trip tables are extracted from the full statewide GM trip table. These "partial" GM trip tables are then merged and a partial GM TLF is created. The GM friction factor curves are adjusted until the partial GM TLF matches the 00 TLF. Three friction factor curves, one for each trip type, resulting from the micro-scale calibration produce a reasonable GM truck trip model. A key methodological issue for GM. calibration involves the use of multiple friction factor curves versus a single friction factor curve for each trip type in order to estimate truck trips with reasonable accuracy. A single friction factor curve for each of the three trip types was found to reproduce the 00 TLFs from the calibration data base. Given the very limited trip generation data available for this research, additional refinement of the gravity model using multiple mction factor curves for each trip type was not warranted. In the traditional urban transportation planning studies, the zonal trip productions and attractions and region-wide OD TLFs are available. However, for this research, the information available for the development .of the GM model is limited to Ground Counts (GC) and a limited set ofOD TLFs. The GM is calibrated using the limited OD data, but the OD data are not adequate to obtain good estimates of truck trip productions and attractions .. Consequently, zonal productions and attractions are estimated using zonal population as a first approximation. Then, Selected Link based (SELINK) analyses are used to adjust the productions and attractions and possibly recalibrate the GM. The SELINK adjustment process involves identifying the origins and destinations of all truck trips that are assigned to a specified "selected link" as the result of a standard traffic assignment. A link adjustment factor is computed as the ratio of the actual volume for the link (ground count) to the total assigned volume. This link adjustment factor is then applied to all of the origin and destination zones of the trips using that "selected link". Selected link based analyses are conducted by using both 16 selected links and 32 selected links. The result of SELINK analysis by u~ing 32 selected links provides the least %RMSE in the screenline volume analysis. In addition, the stability of the GM truck estimating model is preserved by using 32 selected links with three SELINK adjustments, that is, the GM remains calibrated despite substantial changes in the input productions and attractions. The coverage of zones provided by 32 selected links is satisfactory. Increasing the number of repetitions beyond four is not reasonable because the stability of GM model in reproducing the OD TLF reaches its limits. The total volume of truck traffic captured by 32 selected links is 107% of total trip productions. But more importantly, ~ELINK adjustment factors for all of the zones can be computed. Evaluation of the travel demand model resulting from the SELINK adjustments is conducted by using screenline volume analysis, functional class and route specific volume analysis, area specific volume analysis, production and attraction analysis, and Vehicle Miles of Travel (VMT) analysis. Screenline volume analysis by using four screenlines with 28 check points are used for evaluation of the adequacy of the overall model. The total trucks crossing the screenlines are compared to the ground count totals. L V/GC ratios of 0.958 by using 32 selected links and 1.001 by using 16 selected links are obtained. The %RM:SE for the four screenlines is inversely proportional to the average ground count totals by screenline .. The magnitude of %RM:SE for the four screenlines resulting from the fourth and last GM run by using 32 and 16 selected links is 22% and 31 % respectively. These results are similar to the overall %RMSE achieved for the 32 and 16 selected links themselves of 19% and 33% respectively. This implies that the SELINICanalysis results are reasonable for all sections of the state.Functional class and route specific volume analysis is possible by using the available 154 classification count check points. The truck traffic crossing the Interstate highways (ISH) with 37 check points, the US highways (USH) with 50 check points, and the State highways (STH) with 67 check points is compared to the actual ground count totals. The magnitude of the overall link volume to ground count ratio by route does not provide any specific pattern of over or underestimate. However, the %R11SE for the ISH shows the least value while that for the STH shows the largest value. This pattern is consistent with the screenline analysis and the overall relationship between %RMSE and ground count volume groups. Area specific volume analysis provides another broad statewide measure of the performance of the overall model. The truck traffic in the North area with 26 check points, the West area with 36 check points, the East area with 29 check points, and the South area with 64 check points are compared to the actual ground count totals. The four areas show similar results. No specific patterns in the L V/GC ratio by area are found. In addition, the %RMSE is computed for each of the four areas. The %RMSEs for the North, West, East, and South areas are 92%, 49%, 27%, and 35% respectively, whereas, the average ground counts are 481, 1383, 1532, and 3154 respectively. As for the screenline and volume range analyses, the %RMSE is inversely related to average link volume. 'The SELINK adjustments of productions and attractions resulted in a very substantial reduction in the total in-state zonal productions and attractions. The initial in-state zonal trip generation model can now be revised with a new trip production's trip rate (total adjusted productions/total population) and a new trip attraction's trip rate. Revised zonal production and attraction adjustment factors can then be developed that only reflect the impact of the SELINK adjustments that cause mcreases or , decreases from the revised zonal estimate of productions and attractions. Analysis of the revised production adjustment factors is conducted by plotting the factors on the state map. The east area of the state including the counties of Brown, Outagamie, Shawano, Wmnebago, Fond du Lac, Marathon shows comparatively large values of the revised adjustment factors. Overall, both small and large values of the revised adjustment factors are scattered around Wisconsin. This suggests that more independent variables beyond just 226; population are needed for the development of the heavy truck trip generation model. More independent variables including zonal employment data (office employees and manufacturing employees) by industry type, zonal private trucks 226; owned and zonal income data which are not available currently should be considered. A plot of frequency distribution of the in-state zones as a function of the revised production and attraction adjustment factors shows the overall " adjustment resulting from the SELINK analysis process. Overall, the revised SELINK adjustments show that the productions for many zones are reduced by, a factor of 0.5 to 0.8 while the productions for ~ relatively few zones are increased by factors from 1.1 to 4 with most of the factors in the 3.0 range. No obvious explanation for the frequency distribution could be found. The revised SELINK adjustments overall appear to be reasonable. The heavy truck VMT analysis is conducted by comparing the 1990 heavy truck VMT that is forecasted by the GM truck forecasting model, 2.975 billions, with the WisDOT computed data. This gives an estimate that is 18.3% less than the WisDOT computation of 3.642 billions of VMT. The WisDOT estimates are based on the sampling the link volumes for USH, 8TH, and CTH. This implies potential error in sampling the average link volume. The WisDOT estimate of heavy truck VMT cannot be tabulated by the three trip types, I-I, I-E ('||'&'||'pound;-I), and E-E. In contrast, the GM forecasting model shows that the proportion ofE-E VMT out of total VMT is 21.24%. In addition, tabulation of heavy truck VMT by route functional class shows that the proportion of truck traffic traversing the freeways and expressways is 76.5%. Only 14.1% of total freeway truck traffic is I-I trips, while 80% of total collector truck traffic is I-I trips. This implies that freeways are traversed mainly by I-E and E-E truck traffic while collectors are used mainly by I-I truck traffic. Other tabulations such as average heavy truck speed by trip type, average travel distance by trip type and the VMT distribution by trip type, route functional class and travel speed are useful information for highway planners to understand the characteristics of statewide heavy truck trip patternS. Heavy truck volumes for the target year 2010 are forecasted by using the GM truck forecasting model. Four scenarios are used. Fo~ better forecasting, ground count- based segment adjustment factors are developed and applied. ISH 90 '||'&'||' 94 and USH 41 are used as example routes. The forecasting results by using the ground count-based segment adjustment factors are satisfactory for long range planning purposes, but additional ground counts would be useful for USH 41. Sensitivity analysis provides estimates of the impacts of the alternative growth rates including information about changes in the trip types using key routes. The network'||'&'||'not;based GMcan easily model scenarios with different rates of growth in rural versus . . urban areas, small versus large cities, and in-state zones versus external stations. cities, and in-state zones versus external stations.
A costume reveals the social characteristics of the era in which it is worn, thus we can say that the history of change of the costume is the history of change of the living culture of the era. Since the Three States era, the costume structure of this country had been affected by the costume system of the China's historical dynasties in the form of the grant therefrom because of geographical conditions, which affection was conspicuous for the bureaucrat class, particularly including but not limited to the Kings' familities. Such a grant of the costume for the bureaucrat class (i.e., official uniform) was first given by the Dang-dynasty at the age of Queen Jinduck, the 28th of the Shilla-dynasty. Since then, the costume for the bureaucrats had consecutively been affected as the ages had gone from the unified Shilla, to the Koryo and to the Yi-dynasty. As the full costumes officially used by government officials (generally called "Baek Gwan") in the Yidynasty, there existed Jo-bok, Gong-bok and Sang-bok. Of such official costumes, Gong-bok was worn at the time of conducting official affairs of the dynasty, making a respectful visit for the expression of thanks or meeting diplomatic missions of foreign countries. It appears no study was made yet with regard to the Gong-bok while the studies on the Jo-bok and the Sangbok were made. Therefore, this article is, by rendering a study and research on the styles of costumes of civil servants' stone images erected at the Kings' tombs of the Yi-dynasty, to help the persons concerned understand the Gong-bok, one of the official costume for Baek Kwan of that age and further purports to specifically identify the styles and changes of the Gong-bok, worn by Baek Gwan during the Yi-dynasty, consisting of the Bok-doo (a hat, four angled and two storied with flat top), Po (gown), Dae (belt), and Hol (small and thin plate which was officially held by the government officials in hand, showing the courtesy to and writing brief memorandums before the King) and Hwa (shoes). For that purpose, I investigated by actually visiting the tombs of the Kings of the Yi-dynasty including the Geonwon-neung, the tomb of the first King Tae-jo and the You-neung, the tomb of the 27th King Soon-jong as well as the tombs of the lawful wives and concubines of various Kings, totalling 29 tombs and made reference to relevant books and records. Pursuant. to this study, of the 29 Kings' tombs the costume styles of civil servants' stone images erected at the 26 Kings' tombs are those of Gong-bok for Baek-gwan of the Yi-dynasty wearing Bok-doo as a hat and Ban-ryeong or Dan-ryenog Po as a gown with Dae, holding Hol in hand and wearing shoes. Other than those of the 26 tombs, the costume styles of the Ryu-neung, the tomb of the Moon-jo who was the first son of 23rd King Soon-jo and given the King's title after he died and of the You-neung, the tomb of the 27th King Soon-jong are those of Jobok with Yang-gwan (a sort of hat having stripes erected, which is different from the Bok-doo), and that of the Hong-neung, the tomb of the 26th King Go-jong shows an exceptional one wearing Yang-gwan and Ban-ryeong Po ; these costume styles other than Gongbok remain as the subject for further study. Gong-bok which is the costume style of civil servants' stone images of most of the Kings' tombs had not been changed in its basic structure for about 500 years of the Yi-dynasty and Koryo categorized by the class of officials pursuant to the color of Po and materials of Dae and Hol. Summary of this costume style follows: (1) Gwan-mo (hat). The Gwan-mo style of civil servants' stone images of the 26 Kings' tombs, other than Ryu-neung, Hong-neung and You-neung which have Yang-gwan, out of the 29 Kings' tombs of the Yi-dynasty reveals the Bok-doo with four angled top, having fore-part and back-part divided. Back part of the Bok-doo is double the fore-part in height. The expression of the Gak (wings of the Bokdoo) varies: the Gyo-gak Bok-doo in that the Gaks, roundly arisen to the direction of the top, are clossed each other (tombs of the Kings Tae-jong), the downward style Jeon-gak Bok-doo in that soft Gaks are hanged on the shoulders (tombs of the Kings Joong-jong and Seong-jong) and another types of Jeon-gak Bok-doo having Gaks which arearisen steeply or roundly to the direction of top and the end of which are treated in a rounded or straight line form. At the lower edge one protrusive line distinctly reveals. Exceptionally, there reveals 11 Yang-gwan (gwan having 11 stripes erected) at the Ryu-neung of the King Moon-jo, 9 Yang-gwan at the Hong-neung of the King Go-jong and 11 Yang-gwan at the You-neung of the King Soon-jong; noting that the Yang-gwan of Baek Kwan, granted by the Myeong-dynasty of the China during the Yi-dynasty, was in the shape of 5 Yang-gwan for the first Poom (class) based on the principle of "Yideung Chaegang" (gradual degrading for secondary level), the above-mentioned Yang-gwans are very contrary to the principle and I do not touch such issue in this study, leaving for further study. (2) Po (gown). (a) Git (collar). Collar style of Po was the Ban-ryeong (round collar) having small neck-line in the early stage and was changed to the Dan-ryeong (round collar having deep neck-line) in the middle of the: dynasty. In the Dan-ryeong style of the middle era (shown at the tomb of the King Young-jo); a, thin line such as bias is shown around the internal side edge and the width of collar became wide a little. It is particularly noted that the Ryu-neung established in the middle stage and the You-neung in the later stage show civil servants in Jo-bok with the the Jikryeong (straight collar) Po and in case of the Hong-neung, the Hong-neung, the tomb of the King Go-jong, civil servants, although they wear Yang-gwan, are in the Ban-ryeong Po with Hoo-soo (back embroidery) and Dae and wear shoes as used in the Jo-bok style. As I could not make clear the theoretical basis of why the civil servants' costume styles revealed, at these tombs of the Kings are different from those of other tombs, I left this issue for further study. It is also noted that all the civil servants' stone images show the shape of triangled collar which is revealed over the Godae-git of Po. This triangled collar, I believe, would be the collar of the Cheomri which was worn in the middle of the Po and the underwear, (b) Sleeve. The sleeve was in the Gwan-soo (wide sleeve) style. having the width of over 100 centimeter from the early stage to the later stage arid in the Doo-ri sleeve style having the edge slightly rounded and we can recognize that it was the long sleeve in view of block fold shaped protrusive line, expressed on the arms. At the age of the King Young-jo, the sleeve-end became slightly narrow and as a result, the lower line of the sleeve were shaped curved. We can see another shape of narrow sleeve inside the wide sleeve-end, which should be the sleeve of the Cheom-ri worn under the Gong-bok. (c) Moo. The Moo revealed on the Po of civil servants' stone images at the age of the King Sook-jong' coming to the middle era. Initially the top of the Moo was expressed flat but the Moo was gradually changed to the triangled shape with the acute top. In certain cases, top or lower part of the Moo are not reveald because of wear and tear. (d) Yeomim. Yeomim (folding) of the Po was first expressed on civil servants' stone images of the Won-neung, the tomb of the King Young-jo and we can seemore delicate expression of the Yeomim and Goreum (stripe folding and fixing the lapel of the Po) at the tomb of the Jeongseong-wanghoo, the wife of the King Young-jo, At the age of the King Soon-jo, we can see the shape of Goreum similar to a string rather than the Goreum and the upper part of the Goreum which fixes Yeomim was expressed on the right sleeve. (3) Dae. Dae fixed on the Po was placed half of the length of Po from the shoulders in the early stage. Thereafter, at the age of the King Hyeon-jong it was shown on the slightly upper part. placed around one third of the length of Po. With regard to the design of Dae, all the civil servants' stone images of the Kings' tombs other than those of the Geonwon-neung of the King Tae-jo show single or double protrusive line expressed at the edge of Dae and in the middle of such lines, cloud pattern, dangcho (a grass) pattern, chrysanthemum pattern or other various types of flowery patterns were designed. Remaining portion of the waist Dae was hanged up on the back, which was initially expressed as directed from the left to the right but thereafter expressed. without orderly fashion,. to the direction of the left from the right and vice versa, Dae was in the shape of Yaja Dae. In this regard, an issue of when or where such a disorderly fashion of the direction of the remaining portion of waist Dae was originated is also presented to be clarified. In case of the Ryuneung, Hong-neung and You-neung which have civil servants' stone images wearing exceptional costume (Jo-bok), waist Dae of the Ryu-neung and Hong-neung are designed in the mixture of dual cranes pattern, cosecutive beaded pattern and chrvsenthemum pattern and that of You-neung is designed in cloud pattern. (4) Hol. Although materials of the Hol held in hand of civil servants' stone images are not identifiable, those should be the ivory Hol as all the Baek Gwan's erected as stone images should be high class officials. In the styles, no significant changes were found, however the Hol's expressed on civil servants' stone images of the Yi-dynasty were shaped in round top and angled bottom or round top and bottom. Parcicularly, at the age of the King Young-jo the Hol was expressed in the peculiar type with four angles all cut off. (5) Hwa (shoes). As the shoes expressed on civil servants' stone images are covered with the lower edges of the Po, the styles thereof are not exactly identifiable. However, reading the statement "black leather shoes for the first class (1 Poom) to ninth class (9 Poom)," recorded in the Gyeongkook Daejon, we can believe that the shoes were worn. As the age went on, the front tips of the shoes were soared and particularly, at the Hong-neung of the King Go-jong the shoes were obviously expressed with modern sense as the country were civilized.