• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.14 no.4
• /
• pp.332-339
• /
• 2014

Face classification has wide applications in security and surveillance. However, this technique presents various challenges caused by pose, illumination, and expression changes. Face recognition with long-distance images involves additional challenges, owing to focusing problems and motion blurring. Multiple frames under varying spatial or temporal settings can acquire additional information, which can be used to achieve improved classification performance. This study investigates the effectiveness of multi-frame decision-level fusion with photon-counting linear discriminant analysis. Multiple frames generate multiple scores for each class. The fusion process comprises three stages: score normalization, score validation, and score combination. Candidate scores are selected during the score validation process, after the scores are normalized. The score validation process removes bad scores that can degrade the final output. The selected candidate scores are combined using one of the following fusion rules: maximum, averaging, and majority voting. Degraded facial images are employed to demonstrate the robustness of multi-frame decision-level fusion in harsh environments. Out-of-focus and motion blurring point-spread functions are applied to the test images, to simulate long-distance acquisition. Experimental results with three facial data sets indicate the efficiency of the proposed decision-level fusion scheme.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.42 no.4
• /
• pp.316-321
• /
• 2009

This study was conducted to improve the fish odor of extracts obtained from salmon frame. Salmon frame extracts were prepared using four kinds of pretreated salmon frame (salmon frame soaked in soybean milk and fried salmon frame) or containing additives (cystine and xylose-added salmon frame, and methionine and xylose-added salmon frame). Among the extracts prepared in this study, extracts containing cystine and xylose had the highest volatile component intensity and odor sensory score. These suggested that the fish odor of salmon frame extracts can be reduced by adding cystine and xylose before extraction.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.36 no.12
• /
• pp.1596-1603
• /
• 2007

This study was conducted to improve functional properties of salmon frame extracts using various commercial enzymes (Alkalase 2.4 L FG, Flavourzyme 500 MG, Neutrase 0.8 L and Protamex 1.5 MG). The ACE (angiotensin I converting enzyme) inhibitory activity was the highest ($IC_{50}=0.67mg/mL$) in the product incubated with Neutrase for 4 hrs (N4-treated hydrolysates) among the various extracts incubated with commercial enzymes for different times. However, antioxidant activities of all salmon frame extracts were less than 15%. There were no significant differences in the proximate composition and sensory evaluation of the fish odor and taste. However, N4-treated hydrolysate was improved in the extractive-nitrogen content and transmission compared to the other enzymatic hydrolysates. When compared to commercial Gomtang products, N4-treated hydrolysate was also high in protein, extractive-nitrogen, total amino acid, and calcium contents, while low in taste sensory score. There were no differences in transmission and sensory score on the fish odor between N4-treated hydrolysates and commercial Gomtang.

• Phonetics and Speech Sciences
• /
• v.11 no.4
• /
• pp.89-96
• /
• 2019

In this paper, we propose a deep learning architecture that can effectively detect speech segmentation in broadcast contents. We also propose a multi-scale time-dilated layer for learning the temporal changes of feature vectors. We implement several comparison models to verify the performance of proposed model and calculated the frame-by-frame F-score, precision, and recall. Both the proposed model and the comparison model are trained with the same training data, and we train the model using 32 hours of Korean broadcast data which is composed of various genres (drama, news, documentary, and so on). Our proposed model shows the best performance with F-score 91.7% in Korean broadcast data. The British and Spanish broadcast data also show the highest performance with F-score 87.9% and 92.6%. As a result, our proposed model can contribute to the improvement of performance of speech detection by learning the temporal changes of the feature vectors.

• International Journal of Internet, Broadcasting and Communication
• /
• v.11 no.4
• /
• pp.26-30
• /
• 2019

This paper proposes a new voice activity detection (VAD) method which is based on SNR and non-intrusive speech intelligibility estimation. In the conventional SNR-based VAD methods, voice activity probability is obtained by estimating frame-wise SNR at each spectral component. However these methods lack performance in various noisy environments. We devise a hybrid VAD method that uses non-intrusive speech intelligibility estimation as well as SNR estimation, where the speech intelligibility score is estimated based on deep neural network. In order to train model parameters of deep neural network, we use MFCC vector and the intrusive speech intelligibility score, STOI (Short-Time Objective Intelligent Measure), as input and output, respectively. We developed speech presence measure to classify each noisy frame as voice or non-voice by calculating the weighted average of the estimated STOI value and the conventional SNR-based VAD value at each frame. Experimental results show that the proposed method has better performance than the conventional VAD method in various noisy environments, especially when the SNR is very low.

• Journal of Korea Multimedia Society
• /
• v.5 no.5
• /
• pp.512-522
• /
• 2002

In this paper, we propose a hybrid method which is mixed with frame selection and weighting model rank method, based on GMM(gaussian mixture model), for real-time text-independent speaker recognition system. In the system, maximum likelihood estimation was used for GMM parameter optimization, and maximum likelihood was used for recognition basically Proposed hybrid method has two steps. First, likelihood score was calculated with speaker models and test data at frame level, and the difference is calculated between the biggest likelihood value and second. And then, the frame is selected if the difference is bigger than threshold. The second, instead of calculated likelihood, weighting value is used for calculating total score at each selected frame. Cepstrum coefficient and regressive coefficient were used as feature parameters, and the database for test and training consists of several data which are collected at different time, and data for experience are selected randomly In experiments, we applied each method to baseline system, and tested. In speaker recognition experiments, proposed hybrid method has an average of 4% higher recognition accuracy than frame selection method and 1% higher than W method, implying the effectiveness of it.

• Journal of the Korean Home Economics Association
• /
• v.40 no.5
• /
• pp.15-24
• /
• 2002

The following research conclusions were made, relative to the experiments of the textiles of fire fighters Protective Clothing. 1. When the body protection efficiency such as the thickness, the strength and heat resistance are considered, Nomex(N) is tuned out the best outer shelf, Gore-tex(KG) the best moisture barrier, and Wool-felt(WC) the best thermal barrier. 2. In the hygienic and sanitary efficiency also, N is turned out the best outer shelf, KG the best moisture barrier, and WC the best thermal barrier in its degree of water resistance, water vapour permeability, and air permeability. 3. In the washing and maintenance efficiency, too. N is turned out the best outer shell, KG the best moisture barrier, and WC the best thermal barrier, being considered the material's rate of contraction, the changing rate of frame resistance, water resistance, and water vapour permeability. 4. When considered the frame resistance against the reflection tape and reflection efficiency, O is the best material for it marks the highest score in the frame resistance and reflective effect.

• Asian-Australasian Journal of Animal Sciences
• /
• v.28 no.1
• /
• pp.69-78
• /
• 2015

A non-traditional forage-based protocol was employed to evaluate replacement heifer growth, fertility, and economics between small frame (SF, 3.50; n = 50) and large frame (LF, 5.56; n = 50) heifers using three increasing gain growth phases. Preceding an 85 d growing-breeding period (Phase 3; P3) the heifers were managed as a common group for Phases 1 and 2 (P1 and P2). During P1, heifers grazed common fields of unharvested corn and corn residue (total digestible nutrients [TDN] 56%) with supplemental hay. For P2, heifers grazed early spring crested wheatgrass pasture (CWG; TDN 62%) that was followed by the final P3 drylot growing and breeding period (TDN 68%). Small frame heifers were lighter at the end of P1 in May and at the start of P3 breeding in August (p = 0.0002). Percent of mature body weight (BW) at the end of P1 (209 d) was 48.7% and 46.8%, respectively, for the SF and LF heifers and the percent pubertal was lower for SF than for LF heifers (18.0% vs 40.0%; p = 0.02). At breeding initiation (P3), the percentage of mature BW was 57.8 and 57.2 and the percentage pubertal was 90.0 and 96.0 (p = 0.07) for the SF and LF heifers, respectively; a 5-fold increase for SF heifers. Breeding cycle pregnancy on days 21, 42, and 63, and total percent pregnant did not differ (p>0.10). In drylot, SF heifer dry matter intake (DMI) was 20.1% less (p = 0.001) and feed cost/d was 20.3% lower (p = 0.001), but feed cost/kg of gain did not differ between SF and LF heifers (p = 0.41). Economically important live animal measurements for muscling were measured in May and at the end of the study in October. SF heifers had greater L. dorsi muscle area per unit of BW than LF heifers (p = 0.03). Small frame heifer value was lower at weaning (p = 0.005) and the non-pregnant ending heifer value was lower for SF heifers than for the LF heifers (p = 0.005). However, the total development cost was lower for SF heifers (p = 0.001) and the net cost per pregnant heifer, after accounting for the sale of non-pregnant heifers, was lower for SF heifers (p = 0.004). These data suggest that high breeding efficiency can be attained among March-April born SF and LF virgin heifers when transitioned to a more favorable May-June calving period through the strategic use of grazed and harvested forages resulting in a lower net cost per pregnant SF heifer.

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.8
• /
• pp.735-743
• /
• 2002

In this paper, we propose three new text-independent speaker identification methods. At first, to exclude the frames not having enough features of speaker's vocal from calculation of the maximum likelihood, we propose the FS(Frame Selection) method. This approach selects the important frames by evaluating the difference between the biggest likelihood and the second in each frame, and uses only the frames in calculating the score of likelihood. Our secondly proposed, called the Hybrid, is a combined version of the FS and WMR(Weighting Model Rank). This method determines the claimed speaker using exponential function weights, instead of likelihood itself, only on the selected frames obtained from the FS method. The last proposed, called MWMR (Modified WMR), considers both original likelihood itself and its relative position, when the claimed speaker is determined. It is different from the WMR that take into account only the relative position of likelihood. Through the experiments of the speaker identification, we show that the all the proposed have higher identification rates than the ML. In addition, the Hybrid and MWMR have higher identification rate about 2% and about 3% than WMR, respectively.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.8
• /
• pp.2053-2067
• /
• 2023

This paper proposes two video quality assessment methods based on deep neural network. (i)The first method uses the IQF-CNN (convolution neural network based on image quality features) to build image quality assessment method. The LIVE image database is used to test this method, the experiment show that it is effective. Therefore, this method is extended to the video quality assessment. At first every image frame of video is predicted, next the relationship between different image frames are analyzed by the hysteresis function and different window function to improve the accuracy of video quality assessment. (ii)The second method proposes a video quality assessment method based on convolution neural network (CNN) and gated circular unit network (GRU). First, the spatial features of video frames are extracted using CNN network, next the temporal features of the video frame using GRU network. Finally the extracted temporal and spatial features are analyzed by full connection layer of CNN network to obtain the video quality assessment score. All the above proposed methods are verified on the video databases, and compared with other methods.