• Journal of Information Processing Systems
• /
• 제19권6호
• /
• pp.745-755
• /
• 2023

The wire rope is an indispensable production machinery in coal mines. It is the main force-bearing equipment of the underground traction system. Accurate detection of wire rope defects and positions exerts an exceedingly crucial role in safe production. The existing defect detection solutions exhibit some deficiencies pertaining to the flexibility, accuracy and real-time performance of wire rope defect detection. To solve the aforementioned problems, this study utilizes the camera to sample the wire rope before the well entry, and proposes an object based on YOLOv5. The surface small-defect detection model realizes the accurate detection of small defects outside the wire rope. The transfer learning method is also introduced to enhance the model accuracy of small sample training. Herein, the enhanced YOLOv5 algorithm effectively enhances the accuracy of target detection and solves the defect detection problem of wire rope utilized in mine, and somewhat avoids accidents occasioned by wire rope damage. After a large number of experiments, it is revealed that in the task of wire rope defect detection, the average correctness rate and the average accuracy rate of the model are significantly enhanced with those before the modification, and that the detection speed can be maintained at a real-time level.

• 제어로봇시스템학회논문지
• /
• 제12권11호
• /
• pp.1156-1161
• /
• 2006

This paper describes a fast image labeling algorithm for the feature extraction of connected components. Labeling the connected regions of a digitized image is a fundamental computation in image analysis and machine vision, with a large number of application that can be found in various literature. This algorithm is designed for the surface defect inspection of Cold Mill Strip. The labeling algorithm permits to separate all of the connected components appearing on the Cold Mill Strip.

• 제어로봇시스템학회논문지
• /
• 제3권2호
• /
• pp.179-186
• /
• 1997

In this paper we suggest a development of surface defect inspection algorithms for cold mill strip. The defects which exist in a surface of cold mill strip have a scattering or singular distribution. This paper consists of preprocessing, feature extraction and defect classification. By preprocessing, the binarized defect image is achieved. In this procedure, Top-hit transform, adaptive thresholding, thinning and noise rejection are used. Especially, Top-hit transform using local min/max operation diminishes the effect of bad lighting. In feature extraction, geometric, moment and co-occurrence matrix features are calculated. For the defect classification, multilayer neural network is used. The proposed algorithm showed 15% error rate.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• 제18권4호
• /
• pp.746-757
• /
• 1996

Bony defects may be found as a result of congenital anomalies, traumatic injury, automobile collisions and industrial accidents in the maxillofacial area. Such conditions are often associated with severs functional and esthetic problem. Various surgical procedure has been utilized in attempts to repair and reconstruct bony defects. Bone is a complex, living, constantly changing tissue. The architecture and composition of cancellous and cortical bone allow the skeleton to perform its essential mechanical functions. Periosteum covers the external surface of bone and consists of two layers : an outer fibrous layer and an inner more cellular and vascular layer. The inner osteogenic layer or cambium layer can form new bone while the outer layer firms part of the insertions of tendons, ligaments and muscles. This study was under taken to evaluate bone healing process on partial defect of calvarial bone with or without periosteum in rat. We made calvarial defects of different size(4mm, 6mm, 8mm) with periosteum or without periosteum in rat to study the effect of defect size on healing process. Control and experimental groups sacrified at 1, 2, 4, 6, 8 weeks, postoperatively. We examed the specimens by gloss findings, light microscophy, and fluorescent microscophy. The results were as follows. 1. Gloss findings: Control groups are larger bony defects than experimental groups after 2 weeks, and than control groups advanced healing of defected bone but experimental groups are lesser after 4, 6 weeks. After 8 weeks, bone defect has not been identified in control and experimental groups. 2. Light microscope: All defects of control groups are larger bony defects than experimental groups after 2 weeks. And than control groups show smaller defect after 4 weeks. After 8 weeks, the control group reveal pin-point sized, hardly identifiable defect space and the experimental group reveal small, but definite defect space. 3. Fluorescent microscope : Each week, new bone formation of control group is very similar to the experimental group. In this study, Osteogenesis of calvarial bone defects with periosteum or without periosteum was examined for 8 weeks in rats. The replaced periosteum had batter new bone formation than the removed periosteum.

• Journal of Periodontal and Implant Science
• /
• 제29권2호
• /
• pp.355-373
• /
• 1999

Many researches have been reported that collagen as cellular stroma, matrix of grafting materials, mediator of agents for the purpose of promoting healing process invivo, but the responses in vivo were seen various. The goal of this experiment is to assess the effect of collagen on bony healing, through histological evaluation of implanted collagen on the calvarial defect in rats. 2-month-old Sprague-Dawley, 24 rats were used and 12 rats assigned to each group of control and test. Defect of 5mm in diameter was made on the calvarial bone with trephine bur. Following thorough saline rinse, defect of control group was left in empty and that of experimental group was filled with fibrillar collagen($COLLATAPE^{(R)}$, COLLA-TEC. INC. U.S.A.) soaked in saline. 3 rats in each group were sacrificed at 3, 7, 14, 21 days after operation respectively, and the tissue blocks were prepared for light microscope with H-E for evaluation of overall healing, with TRAP(tartrate resistant acid phosphatase) for evaluation of osteoclastic activity and with immunohistochemical staining for macrophages. The results were as follows : 1. In the control group, inflammatory responses were disappeared at day 14, but, in the experimental group inflammatory infiltrates were reduced at day 21. Thus, the experimental group showed more severe soft tissue inflammation than control group. 2. Both control and experimental group showed slight appositional growth at day 7 and gradual bony growth to 21th day. But, complete bony healing of the defect was not shown. There was no significant difference in bony healing between control and experimental group 3. Specific response of macrophages for implanted collagen was observed at day 14 in the experimental group. In conclusion, although fibrillar collagen caused inflammation of soft tissue during initial healing period, inflammatory responses by fibrillar collagen didn't inhibit bony regeneration and implanted collagen was biodegradaded by macrophages. Thus, we expect that fibrillar collagen can be used for useful mediator of graft materials or growth factors.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• 제37권
• /
• pp.32.1-32.5
• /
• 2015

Background: The aims of present study were (1) to evaluate new bone formation among the 4-hexylresorcinol (4HR)-incorporated silk fabric membrane (SFM), conventional SFM, and uncovered control groups and (2) to compare the amount of residual membrane between the 4HR-incorporated SFM and conventional SFM in a rabbit parietal defect model. Methods: Nine New Zealand white rabbits were used for this animal study. After the formation of a bilateral parietal bone defect (diameter 8.0 mm), either 4HR-incorporated SFM or conventional SFM was grafted into the defect. The defect in the control was left uncovered. New bone formation and the amount of residual membrane were evaluated by histomorphometry at 8 weeks after the operation. Results: The total amount of new bone was $37.84{\pm}8.30%$ in the control, $56.64{\pm}15.74%$ in the 4HR-incorporated SFM group, and $53.35{\pm}10.52%$ in the conventional SFM group 8 weeks after the operation. The differences were significant between the control and 4HR-incorporated SFM group (P = 0.016) and between the control and conventional SFM group (P = 0.040). The residual membrane was $75.08{\pm}10.52%$ in the 4HR-incorporated SFM group and $92.23{\pm}5.46%$ in the conventional SFM group 8 weeks after the operation. The difference was significant (P = 0.039). Conclusions: The 4HR-incorporated SFM and conventional SFM groups showed more bone regeneration than the control group. The incorporated 4HR accelerated the partial degradation of the silk fabric membrane in a rabbit parietal defect model 8 weeks after the operation.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 2009년도 추계학술대회 논문집
• /
• pp.169-172
• /
• 2009

Multiple scale modeling has been applied to predict defect shape change during the wire rod rolling process. The size difference between bloom and defect prevent using usual FEM approaches due to the enormous number of elements required to depict the defect. The newly developed multiple scale model can visualize defect shape changes during the multi stands rolling process. The defect positioned at the top and side of bloom are smoothed out but the one at the middle evolved as folding or remained as crack. This approach can be used for defect control with roll shape design and initial bloom shape.

• Journal of Mechanical Science and Technology
• /
• 제18권9호
• /
• pp.1493-1499
• /
• 2004

This paper describes the speckle shearing interferometry, a non-destructive optical method, for quantitative estimation of void defect and monitoring separation defect inside of a pneumatic tire. Previous shearing interferometry has not supplied quantitative result of inside defect, due to effective factors. In the study, factors related to the details of an inside defect are classified and optimized with pipeline simulator. The size and the shape of defect can be estimated accurately to find a critical point and also is closely related with shearing direction. The technique is applied for quantitative estimation of defects inside of a pneumatic tire. The actual traveling tire is monitored to reveal the cause of separation and the starting points. And also unknown void defects on tread are inspected and the size and shape of defects are estimated which has good agreement with the result of visual inspection.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 2000년도 춘계학술대회논문집
• /
• pp.155-158
• /
• 2000

Flow defect of a piston-pin for automobile parts is investigated in this study. In cold forging of piston-pin Lapping defect a kind of flow defect appears by the dead metal zone. This appearance evidently happens in products with a thin piercing thickness for the dimension accuracy and the decrease of material loss. The best method that can prevent flow defect is removing dead metal zone. The finite element simulations are applied to analyze the flow defect. This study proposed processes for preventing flow defect by removing dead metal zone. Then the results are compared with the experiments for verification. These FE simulation results are in good agreement with the experimental ones.

• Journal of Information Processing Systems
• /
• 제19권1호
• /
• pp.109-117
• /
• 2023

Defect detection is one of the key factors in fabric quality control. To improve the speed and accuracy of denim fabric defect detection, this paper proposes a defect detection algorithm based on cascading feature extraction architecture. Firstly, this paper extracts these weight parameters of the pre-trained VGG16 model on the large dataset ImageNet and uses its portability to train the defect detection classifier and the defect recognition classifier respectively. Secondly, retraining and adjusting partial weight parameters of the convolution layer were retrained and adjusted from of these two training models on the high-definition fabric defect dataset. The last step is merging these two models to get the defect detection algorithm based on cascading architecture. Then there are two comparative experiments between this improved defect detection algorithm and other feature extraction methods, such as VGG16, ResNet-50, and Xception. The results of experiments show that the defect detection accuracy of this defect detection algorithm can reach 94.3% and the speed is also increased by 1-3 percentage points.