• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.9C
• /
• pp.861-866
• /
• 2007

A clustering algorithm for Gaussian Probability Distribution Function (GPDF) data called Centroid Neural Network with a Bhattacharyya Kernel (BK-CNN) is proposed in this paper. The proposed BK-CNN is based on the unsupervised competitive Centroid Neural Network (CNN) and employs a kernel method for data projection. The kernel method adopted in the proposed BK-CNN is used to project data from the low dimensional input feature space into higher dimensional feature space so as the nonlinear problems associated with input space can be solved linearly in the feature space. In order to cluster the GPDF data, the Bhattacharyya kernel is used to measure the distance between two probability distributions for data projection. With the incorporation of the kernel method, the proposed BK-CNN is capable of dealing with nonlinear separation boundaries and can successfully allocate more code vector in the region that GPDF data are densely distributed. When applied to GPDF data in an image classification probleml, the experiment results show that the proposed BK-CNN algorithm gives 1.7%-4.3% improvements in average classification accuracy over other conventional algorithm such as k-means, Self-Organizing Map (SOM) and CNN algorithms with a Bhattacharyya distance, classed as Bk-Means, B-SOM, B-CNN algorithms.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.2
• /
• pp.375-383
• /
• 2012

The purpose of the ensemble methods is to increase the accuracy of prediction through combining many classifiers. According to recent studies, it is proved that random forests and forward stagewise regression have good accuracies in classification problems. However they have great prediction error in separation boundary points because they used decision tree as a base learner. In this study, we use the kernel ridge regression instead of the decision trees in random forests and boosting. The usefulness of our proposed ensemble methods was shown by the simulation results of the prostate cancer and the Boston housing data.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.63 no.2
• /
• pp.131-139
• /
• 2018

Unsaponifiables in the kernel and the cob of 7 maize varieties were analyzed by using thin-layer chromatography (TLC) and gas chromatography (GC) for the identification of phytosterols and their concentrations. The unsaponifiables of the kernel were clearly separated into band I (campesterol, stigmasterol, and ${\beta}$-sitosterol), band II (${\Delta}^5$-avenasterol), band III (${\Delta}^7$- stigmastenol), and band IV (${\Delta}^7$-avenasterol). In the cob, on the other hand, three or more bands were separated in addition to bands. The GC analysis of unsaponifiables showed good separation of campesterol, stigmasterol and ${\beta}$-sitosterol, but the mixture of ${\Delta}^7$-avenasterol (retention time[RT] 22.846), ${\Delta}^7$-stigmastenol (RT 22.852), and ${\Delta}^5$-avenasterol (RT 22.862) showed poor separation. Phytosterol content of the maize kernel was 635.9 mg/100 g, and that of the cob was 273.0 mg/100 g, respectively. The phytosterol content of the kernel was 2.4-fold higher than that of the cob. The phytosterol content of the kernel was higher in the order ${\beta}$sitosterol 80.05% > campesterol 10.5% > stigmasterol 9.46%, but that of the cob was higher in the order ${\beta}$-sitosterol 59.43% > stigmasterol 31.72% > campesterol 10.98%. Based on these results, it appears that the phytosterols of the maize kernel are synthesized in the maize cob and are transferred to the kernel, because the precursors (${\Delta}^7$-avenasterol, ${\Delta}^7$-stigmastenol, and ${\Delta}^5$-avenasterol) of major phytosterols were detected in maize cobs.

• Journal of Biosystems Engineering
• /
• v.34 no.1
• /
• pp.21-29
• /
• 2009

A rapeseed reaping equipment attachable to a conventional combine was developed in order to harvest rapeseed for bio-diesel materials. This study was carried out to measure the harvest feasibility of a prototype combine in rapeseed fields. Grain, stem and pod flow rate, grain qualities (whole kernel, damaged kernel, unhulled kernel, material-other-than-grain) and grain loss rates (header, threshing, separation) were investigated in each field test. As the result of the fold test, the average grain flow rates of SUNMANG and MS varieties showed 1,430 kg/h and 2,038 kg/h, respectively. The average stem and pod flow rates showed 3,443 kg/h and 6,596 kg/h, respectively. In each working speed, the average whole kernel rate and the material-other-than-grain showed 99.9% and below 0.08%, respectively. In the average grain loss, the rates showed 5.66% in case of SUNMANG and 5.94% in MS. Header loss was higher than other parts for SUNMANG. However, threshing loss was relatively higher than other parts for MS. Header loss rate due to side cutter knifes, however, was not so high when compared with a grain loss due to the cutter bar. Effective field capacity and field efficiency of the prototype combine showed 0.389 ha/h and 44%, respectively. Comparison of customary combine with the prototype combine through field test demonstrated that the header loss was reduced by 69.3% when the prototype combine was used.

• Journal of Biosystems Engineering
• /
• v.13 no.3
• /
• pp.32-43
• /
• 1988

This study was attempted to investigate the pneumatic separation on separating unit of a combine harvester. The aerodynamic characteristics of threshed materials were analyzed by experiments. The air velocity distribution within the separation chamber was measured for various speeds of the winnower and suction fans to find out the operational and design conditions of the separating unit which would serve for reducing the grain loss from chaff outlet. The results of study arc summarized as follows: 1. Based on the separation curve of threshed materials analyzed, it was shown that three different kind. of materials-kernels, straw chaff, and leaf chaff were as a whole able to be separated pneumatically, regardless of varieties. However, a small amount of the separation grain loss may be expected to occur if the complete separation between kernels and straw chaff would be undertaken because some portion of their separation curve were overlapping. 2. The analysis of air velocity distribution showed that the separation chamber may be divided into two regions, the discharging and separating. The air velocity of the discharging region was 5-15 m/s and that of the separating region 2-5 m/s. 3. The air movement of the separation chamber may be a turbulence flow, being its speed became greater as it moves from the left to the right section of the separation chamber. The equi-speed line. of air flow had a steep gradient in between the discharging and the separation regions. The air velocity in the discharging region was much higher than the terminal velocity of kernels, because of which those kernels appearing in the region could be possibly exhausted as the grain loss from the chaff outlet. 4. The motion trajectory of threshed material in the separating region was dominantly affected by the winnower fan, on the other hand, its motion in the discharging region was affected by suction fan. 5. The grain loss from the chaff outlet was affected greatly by the winnower fan and the trace of kernel movement. It was observed that the optimum working speed to give minimum grain loss from chaff outlet for the combine tested should be maintained at 950~1,150 rpm for the winnower fan and 1,850 rpm for the suction fan. 6. It was shown that a large portion of grain loss from chaff outlet may occur when the kernels may bump against a portion of separation chamber wall and those kernels thus scattered into the discharging region were sucked by the suction fan. It was accordingly recommended that a new design of the wall of separation chamber so as to bump down kernels may be necessary to reduce grain loss from the chaff outlet.

• Journal of Biosystems Engineering
• /
• v.38 no.2
• /
• pp.95-102
• /
• 2013

Purpose: Using a laboratory indented cylinder separator, broken rice separation experiments were conducted and the characteristics of the separation process were studied to provide information for developing a prototype indented cylinder broken rice separator. Methods: Rice (Ilmi variety) milled in a local RPC was used for the experiment. Rice kernels were classified into four groups according to their length l; whole kernels (I > 3.75 mm), semi-whole kernels (2.5 < I < 3.75 mm), broken kernels (1.75 < I < 2.5 mm), and foreign matters (I < 1.75 mm). A laboratory grain cleaner, Labofix '90 (Schmidt AG, Germany) was used for the experiments. Experiments were designed as a $4{\times}4$ factorial arrangement in randomized blocks with three replications. Cylinder rotational speeds (17, 34, 51, 68 rpm) and trough angles (15, 37.5, 60, $82.5^{\circ}$) were the two factors and feed rates (25, 50 kg/h), indent shapes (Us, $S_1$ type), and indent sizes (2.5, 3.75 mm) were treated as the blocks. Two 125 g samples and one 125 g sample were taken at the cylinder outlet and from the trough, respectively. The whole, semi-whole, and broken kernel weight ratio of the samples and feed was determined by a rice sizing device. From these weight ratios, purities, degrees of extraction and coefficient of separation efficiency were calculated. Results: Trough angle, cylinder speed, and their interaction on the coefficient of separation efficiency were statistically significant. Cylinder speed of 17, 34, and 51 rpm made the most effective separation when the trough angle was $15^{\circ}$ or $37.5^{\circ}$, $60^{\circ}$, and $82.5^{\circ}$, respectively. Maximum values of coefficient of separation efficiency were in the range of 60 to 70% except when the indent size was 2.5 mm and were recorded for the combinations of low cylinder speed (17 rpm) with medium trough angle ($37.5^{\circ}$ or $60^{\circ}$). Indent shape did not appear to make any noticeable difference in separation efficiency. Conclusions: Due to the interaction effect, the trough angle needs to be increased appropriately when an increase in cylinder speed is made if a rapid drop of effectiveness of separation should be avoided. In commercial applications, $S_1$ type indents are preferred because of their better manufacturability and easier maintenance. For successful separation of broken kernels, the indent size should be set slightly bigger than the actual sizes of broken kernels: an indent size of 3.0 mm for separating broken kernels shorter than 2.5 mm.

• Asian-Australasian Journal of Animal Sciences
• /
• v.13 no.6
• /
• pp.861-882
• /
• 2000

The seed, or grain, of modern cultivars of Lupinus angustifolius, commonly known as Australian sweet lupins (ASL), is an established feed resource for the intensive animal industries of Australia, Japan, Korea and several other countries in Asia and Europe. Since the introduction of ASL to the world marketplace about 25 years ago, researchers in many countries have found them to be a valuable component of the diet of beef and dairy cattle, sheep, pigs, poultry, finfish and crustaceans. The seed of ASL contains ~32% crude protein (CP) (~35% DM basis) and 5% oil. The main storage carbohydrates in the seed are the ${\beta}$-galactans that comprise most of the cell-wall material of the kernel and the cellulose and hemicellulose of the thick seed coats. ASL seeds contain about 40% non-starch polysaccharides (NSP) and a negligible amount of starch. This makes them an excellent ingredient for ruminant diets, as the risk of acidosis is very low. The seed of modern cultivars of domesticated Lupinus species contain negligible amounts of lectins and trypsin inhibitors so they do not require preheating before being used as an ingredient in feeds for monogastric species. They have a high digestibility coefficient for protein, >90% for most species, but a low energy digestibility, ~60%, which is mostly due to the high content of NSP. The low content of methionine (0.22%) and of lysine (1.46%) is typical of the legumes. The lysine availability for pigs is >70%. Lupin kernels contain ~39% CP (~42% DM basis), 6% oil and 30% NSP. They have a higher digestible energy for pigs and finfish and a higher metabolisable energy for poultry than whole seed. Commercial operations rarely achieve complete separation of kernel from hull and it is more likely that the kernel fraction, called splits or meats, will contain ~36% CP. The replacement of soybean meal or peas with ASL in cereal-based diets for most intensively reared animals, birds and fish is possible provided lysine, methionine and digestible energy levels are kept constant. This makes ASL economically competitive in many, but not all, circumstances.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.5
• /
• pp.33-38
• /
• 2016

In Android platform, as applications invoke various service functions through IPC (Inter-Process Communication), IPC performance is critical to the responsiveness in Android. However, Android offers long IPC latency of hundreds of micro-seconds due to complicated software stacks between the kernel Binder and the user-level process Context Manager. This separation provides modularity and flexibility, but degrades the responsiveness of services owing to additional context switching and inefficient request handling. In this paper, we anatomize Android IPC mechanisms and observe that 55% of IPC latency comes from the communication overhead between Binder and Context Manager. Based on this observation, this paper proposes a kernel Binder cache that retains a popular subset of service function mappings, thereby reducing the requests transferred to the user-level daemon. The proposed Binder cache is implemented in Android 5.0 and experimental results with various benchmarks show that the proposed cache architecture improves performance by 52.9% on average.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.1
• /
• pp.125-130
• /
• 2018

Linear discriminant analysis is a statistical analysis method that is generally used for dimensionality reduction of the feature vectors or for class classification. However, in the case of a data set that cannot be linearly separated, it is possible to make a linear separation by mapping a feature vector into a higher dimensional space using a nonlinear function. This method is called generalized discriminant analysis or kernel discriminant analysis. In this paper, we carried out target classification experiments with active sonar target signals available on the Internet using both liner discriminant and generalized discriminant analysis methods. Experimental results are analyzed and compared with discussions. For 104 test data, LDA method has shown correct recognition rate of 73.08%, however, GDA method achieved 95.19% that is also better than the conventional MLP or kernel-based SVM.

• 제어로봇시스템학회:학술대회논문집
• /
• 1998.10a
• /
• pp.271-274
• /
• 1998

This paper describes a new method for separation of the Volterra kernels which are identified by use of M-sequence. One of the authors has proposed a method for identification of Volterra kernels of nonlinear systems using M-sequence and correlation technique. When M-sequence are applied to a nonlinear systems, the cross-correlation function between the input and the output of the nonlinear systems includes cross-sections of high-order Volterra kernels. However, if various order Volterra kernels exixt on the obtained cross-correlation function, it is difficult to separate the Volterra kernels. In this paper, the authors show that the magnitude of Volterra kernels is maginified by the amplitude of M-sequence according to the order of Volterra kernels. By use of this property, each order Volterra kernels is obtained by solving linear equations. Simulations are carried out for some nonlinear systems. The results show that Volterra kernels can be separated in each order successfully by the proposed method.