An Improved Convolutional Neural Network-Based Scene Image Recognition Method

Wang, Pinhe; Qiao, Jianzhong; Liu, Nannan

doi:10.1155/2022/3464984

Cited by 7 publications

(7 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The use of guided gradient-weighted class activation mapping (Grad-CAM) is employed to localize the dose distribution regions strongly correlated with cases categorized as ≥ 2 and < 2 in radiation pneumonitis (RP). Features learned by the convolutional lters are generated through gradient ascent to understand the deep network, and the model exhibits excellent performance [19]. A novel CNN-based object detection model is utilized to strike a better balance between accuracy and speed in recognizing high-speed scene images, demonstrating effective practical results in scene image recognition [20].…”

Section: Discussionmentioning

confidence: 99%

The Combined Improvement of Acute Exercise Fatigue through Hyperbaric Oxygen and Astaxanthin Under the Background of Artificial Intelligence

Zhang,

Liu,

Huang

et al. 2024

Preprint

View full text Add to dashboard Cite

Objective this work explored the application research of artificial intelligent (AI) algorithms in the oral and psychological rehabilitation of patients with chronic diseases through deep learning and hyperbaric oxygen combined with astaxanthin therapy. Materials and Methods a model was constructed using convolutional neural network (CNN) combined with multivariate gaussian distribution (MGD) to comprehensively analyze multi-mode health data (numeric, textual, and image). The performance of the CNN model and the CNN + MGD model was compared in terms of accuracy (Acc), area under the receiver operating characteristic curve (AUC), precision (Pre), and recall rate (Recall). Shanghai active football players were selected as the research subjects and divided into a control group, a hyperbaric oxygen intervention group, and a hyperbaric oxygen + astaxanthin intervention group. An AI system was adopted to analyze the electronic health data of patients, assess the efficacy of hyperbaric oxygen in eliminating acute fatigue in patients with chronic diseases, and explore its relationship with oral and psychological health. Results the Acc, AUC, Pre, and Recall of the CNN + MGD model were significantly higher than those of the three-dimensional CNN model (P < 0.05), demonstrating superior performance. Hyperbaric oxygen + astaxanthin treatment significantly promoted the elimination of acute fatigue in patients (P < 0.05), accelerated the recovery of blood oxygen saturation and muscle oxygen saturation, and significantly affected the degradation of blood lactate (P < 0.05). As a result, it rapidly reduced blood oxygen saturation and muscle oxygen saturation after acute fatigue exercise. The oral health-related quality of life in patients with chronic diseases was positively correlated with depression and anxiety levels. Psychological intervention can be conducted in clinical treatment to improve the treatment effect. Conclusion this work comprehensively evaluated the oral and psychological rehabilitation effects in patients with chronic diseases through deep analysis of electronic health data using intelligent algorithms that integrate deep learning with hyperbaric oxygen and astaxanthin therapy. The results suggested that hyperbaric oxygen + astaxanthin treatment not only significantly improved blood indicators but also exhibited a positive correlation with oral health-related quality of life and psychological health, providing a new treatment approach for patient recovery. Therefore, the research results were of great significance for improving patient quality of life and enhancing clinical treatment efficacy.

show abstract

Section: Discussionmentioning

confidence: 99%

The Combined Improvement of Acute Exercise Fatigue through Hyperbaric Oxygen and Astaxanthin Under the Background of Artificial Intelligence

Zhang,

Liu,

Huang

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…When an image is small, with few main features, or when the object in the image is tilted, ordinary CNN may not be able to identify it accurately. Therefore, literature [16] proposes a new CNN object detection model, which improves the algorithm of the CNN model and combines it with the STN model. STN network model is an attention mechanism used to enhance the CNN model.…”

Section: Stn-cnn Composite Modelmentioning

confidence: 99%

Research on image recognition in human-computer interaction based on convolutional neural networks

Yang

2024

ACE

View full text Add to dashboard Cite

Image recognition is an important research direction in human-computer interaction, which has broad development prospects, aiming at enabling computers to understand and interpret image content. Early image recognition methods mainly rely on hand-designed feature extraction algorithms for image analysis and classification. However, this method has significant limitations, which may not provide accurate recognition results for complex images and is hard to adapt to different application scenarios. With the advancement of deep learning and increased processing power in recent years, Convolutional neural network (CNN) -based image recognition methods have achieved remarkable achievements in human-computer interaction. Therefore, based on the image recognition of CNN in human-computer interaction, this paper first studies the CNN model in depth then describes the application of CNN in human-computer interaction, then enumerates different design methods for comparative analysis, and finally summarizes the advantages and disadvantages of CNN in application and proposes improvements to existing problems. The research shows that image recognition based on CNN is better than the traditional network model and has higher accuracy, but it still has some disadvantages. To address these issues and produce more effective and precise image understanding and interaction, it is required to research and enhance the model's structure and algorithm. The appearance and development of CNN have greatly promoted the development of image recognition technology, which has been widely used in human-computer interaction and has made great breakthroughs.

show abstract

“…The use of Convolutional Neural Networks (CNNs) has contributed significantly to health sciences 17 – 19 , especially for image-based diagnosis 20 – 22 . In the last few years, artificial intelligence entered the dental age estimation arena as an alternative to promote automation in the field 23 – 25 .…”

Section: Introductionmentioning

confidence: 99%

Binary decisions of artificial intelligence to classify third molar development around the legal age thresholds of 14, 16 and 18 years

Franco,

Murray,

Heng

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Third molar development is used for dental age estimation when all the other teeth are fully mature. In most medicolegal facilities, dental age estimation is an operator-dependent procedure. During the examination of unaccompanied and undocumented minors, this procedure may lead to binary decisions around age thresholds of legal interest, namely the ages of 14, 16 and 18 years. This study aimed to test the performance of artificial intelligence to classify individuals below and above the legal age thresholds of 14, 16 and 18 years using third molar development. The sample consisted of 11,640 panoramic radiographs (9680 used for training and 1960 used for validation) of males (n = 5400) and females (n = 6240) between 6 and 22.9 years. Computer-based image annotation was performed with V7 software (V7labs, London, UK). The region of interest was the mandibular left third molar (T38) outlined with a semi-automated contour. DenseNet121 was the Convolutional Neural Network (CNN) of choice and was used with Transfer Learning. After Receiver-operating characteristic curves, the area under the curve (AUC) was 0.87 and 0.86 to classify males and females below and above the age of 14, respectively. For the age threshold of 16, the AUC values were 0.88 (males) and 0.83 (females), while for the age of 18, AUC were 0.94 (males) and 0.83 (females). Specificity rates were always between 0.80 and 0.92. Artificial intelligence was able to classify male and females below and above the legal age thresholds of 14, 16 and 18 years with high accuracy.

show abstract

An Improved Convolutional Neural Network-Based Scene Image Recognition Method

Cited by 7 publications

References 20 publications

The Combined Improvement of Acute Exercise Fatigue through Hyperbaric Oxygen and Astaxanthin Under the Background of Artificial Intelligence

The Combined Improvement of Acute Exercise Fatigue through Hyperbaric Oxygen and Astaxanthin Under the Background of Artificial Intelligence

Research on image recognition in human-computer interaction based on convolutional neural networks

Binary decisions of artificial intelligence to classify third molar development around the legal age thresholds of 14, 16 and 18 years

Contact Info

Product

Resources

About