5 results on '"Zhang, Huang"'
Search Results
2. Image inpainting forensics method based on dual branch network
- Author
-
Dengyong ZHANG, Huang WEN, Feng LI, Peng CAO, Lingyun XIANG, Gaobo YANG, Xiangling DING
- Subjects
image forensics ,image forgery detection ,deep learning ,attention mechanism ,Electronic computers. Computer science ,QA75.5-76.95 - Abstract
Image inpainting is a technique that uses information from known areas of an image to repair missing or damaged areas of the image.Image editing software based on it has made it easy to edit and modify the content of digital images without any specialized foundation.When image inpainting techniques are used to maliciously remove the content of an image, it will cause confidence crisis on the real image.Current researches in image inpainting forensics can only effectively detect a certain type of image inpainting.To address this problem, a passive forensic method for image inpainting was proposed, which is based on a two-branch network.The high-pass filtered convolutional network in the dual branch first used a set of high-pass filters to attenuate the low-frequency components in the image.Then features were extracted using four residual blocks, and two transposed convolutions were performed with 4x up-sampling to zoom in on the feature map.And thereafter a 5×5 convolution was used to attenuate the tessellation artifacts from the transposed convolutions to generate a discriminative feature map on the high-frequency components of the image.The dual-attention feature fusion branch in the dual branch first added a local binary pattern feature map to the image using a preprocessing block.Then the dual-attention convolution block was used to adaptively integrate the image’s local features and global dependencies to capture the differences in content and texture between the inpainted and pristine regions of the image.Additionally, the features extracted from the dual-attention convolution block were fused, and the feature maps were up-sampled identically to generate the discriminative image content and texture on the feature maps.The extensive experimental results show the proposed method improved the F1 score by 2.05% and the Intersection over Union(IoU) by 3.53% for the exemplar-based method and by 1.06% and 1.22% for the deep-learning-based method in detecting the inpainted region of the removed object.Visualization of the results shows that the edges of the removed objects can be accurately located on the detected inpainted area.
- Published
- 2022
- Full Text
- View/download PDF
3. Deep Hash Retrieval Algorithm for Medical Images Based on Attention Mechanism
- Author
-
ZHU Cheng-zhang, HUANG Jia-er, XIAO Ya-long, WANG Han, ZOU Bei-ji
- Subjects
medical image retrieval ,attention mechanism ,deep hashing ,bayesian framework ,convolutional neural networks ,Computer software ,QA76.75-76.765 ,Technology (General) ,T1-995 - Abstract
A medical image retrieval method combining attention mechanism is proposed for a series of problems such as poor retrieval performance,low accuracy and lack of interpretability in current medical image retrieval.Based on deep convolutional neural networks and taking Bayesian models as the framework,the proposed algorithm introduces an attention mechanism module guided by semantic features.Local feature descriptors containing semantic information are generated under the guidance of the classification network.Both global features and local features rich in semantic information are used as inputs to the hash network,which enhances the feature representation capability of hash coding by guiding the hash network to pay attention to important feature regions from both global and local perspectives.And the weighted likelihood estimation function is introduced to solve the problem of the unbalanced number of positive and negative sample pairs.MAP and NDCG are used as evaluation metrics,and the ChestX-ray14 dataset is selected for experiments.The proposed algorithm is compared with the current commonly used deep ha-shing methods.Experiment results show that the MAP and NDCG values are much better than the existing deep hashing methods at different code levels of hash coding,which proves the effectiveness of the proposed algorithm.
- Published
- 2022
- Full Text
- View/download PDF
4. Application of retrobulbar nerve block combined with general anesthesia in corneal transplantation
- Author
-
Lai Aihua, Zhang Huang, You Shan, Sun Tao, Liu Xian, and Zhou Wentian
- Subjects
visual analogue scale ,inflammatory cytokine ,retrobulbar nerve block ,propofol ,lcsh:R ,lcsh:Medicine ,surgical stress response ,penetrating keratoplasty ,local anesthesia ,postoperative pain ,hemodynamics ,general anesthesia ,remifentanil - Abstract
Objective To explore the application value of retrobulbar nerve block combined with general anesthesia in the penetrating keratoplasty (PKP). Methods A total of 100 recipients undergoing PKP from January 2017 to January 2019 were recruited in this study. All recipients were divided into the observation group (n=50) and control group (n=50) by random number table method. In the control group, patients received laryngeal mask airway under general anesthesia, and in the observation group, patients received laryngeal mask airway under general anesthesia combined with retrobulbar nerve block. Hemodynamic changes of the PKP recipients before and after operation were observed in two groups. The dosage of analgesic drugs and the incidence of complications were observed in two groups. The degree of pain at postoperative 2-, 6- and 24-h was evaluated by visual analogue scale (VAS) in two groups. The awakening situation of the recipients in two groups was observed. The levels of inflammatory cytokines at 1 d before and after operation were statistically compared in two groups. Results The average arterial pressure and heart rate at intraoperative 15 min and after the surgery in the observation group were significantly higher than those in the control group (both P < 0.05). In the observation group, the dosage of remifentanil and propofol were (1.0±0.4) mg and (299±40) mg, significantly lower than (1.3±0.6) mg and (365±42) mg in the control group (both P < 0.05). The incidence of complications did not significantly differ between two groups (P > 0.05). In the observation group, the VAS scores at 2-, 6- and 12-h after operation were remarkably lower than those in the control group (all P < 0.01). The respiratory recovery time, eye opening time, directional force recovery time and extubation time of the recipients in the observation group were significantly shorter than those in the control group (all P < 0.05). The expression levels of including interleukin (IL)-1, IL-6 and tumor necrosis factor-α (TNF-α) at postoperative 1 d in the observation group were considerably lower than those in the control group (all P < 0.05). Conclusions Retrobulbar nerve block combined with general anesthesia can maintain hemodynamic stability during PKP, reduce the dosage of remifentanil and propofol and alleviate the degree of postoperative pain and inflammatory responses of the recipients.
- Published
- 2019
5. Cryptococcal granuloma of the right frontal lobe and related review of the literature
- Author
-
Zhuo⁃fang HAO, Xiao⁃ming OUYANG, Kai⁃yong MEI, Chun⁃hui ZHOU, and Shi⁃zhang HUANG
- Subjects
Cryptococcosis ,Granuloma ,Frontal lobe ,Immunohistochemistry ,Neurology. Diseases of the nervous system ,RC346-429 - Abstract
Objective To explore the clinicopathological and imaging features of cryptococcal granuloma in central nervous system (CNS), and review related literatures to enhance the recognition and diagnostic level. Methods Retrospectively analysed clinicopathological, histopathological and immunohistochemical features and imaging findings of one case of cryptococcal granuloma in the right frontal lobe. HE staining, alcian blue⁃periodic acid⁃Schiff (AB⁃PAS) staining, Gomori silvermethenamine (periodic acid silvermethenamine, PASM) staining, and immunohistochemistry staining were used respectively. CD68 was determined. The related literatures were reviewed. Results A 30⁃year⁃old male patient presented with sudden limb convulsion and accompanied with temporary loss of consciousness before 6 d. Cranial computed tomography (CT) showed a nodular hyperdense lesion in the right frontal lobe. Cranial magnetic resonance imaging (MRI) examination demonstrated an enhancing mass in the right frontal lobe area, measuring 3.10 cm × 2.80 cm × 1.50 cm in dimension, which showed low T2 ⁃ weighted signal intensity, contrasting with the surrounding hyperintense cerebral edema and irregular lobulated margin of the ring⁃enhancing lesion on T1WI. Grossly, grayish and (or) yellowish gray chunky tissues were found, measuring 2.80 cm × 2.50 cm × 1.50 cm. Pathological finding revealed a large number of cryptococcal lesions and granuloma formation. Final diagnosis was cryptococcal granuloma or cryptococcoma of the right frontal lobe. Conclusion Cryptococcosis is a subacute or chronic fungal disease caused by cryptococcus neoformans, mainly affects CNS and lungs. There are two main forms of infection of the CNS, which include meningitis (with or without parenchymal cysts) and cryptococcal abscess (cryptococcoma). Cryptococcal meningitis and meningoencephalitis are most common. Cryptococcal granuloma is rare. Cerebral cryptococcosis is often misdiagnosed, we should pay close attention to it. Early diagnosis and prompt treatment is crucial. DOI:10.3969/j.issn.1672-6731.2010.04.020
- Published
- 2010
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.