Search or ask a question

How to turn off automatic ear detection for Airpods on Android?

Feature extraction

Haar-like features

Object detection

Answers from top 9 papers

PDF

Open Access

More filters

Papers (9)	Insight
Proceedings Article•DOI Ear Detection Based on Improved AdaBoost Algorithm Wen-Jing Li, Zhi-Chun Mu - Show less +1 more 31 Oct 2008 37 Citations	The third contribution is we designed an ear detection system of DSP and gained a good result of practical application.
Open access•Proceedings Article•DOI A Skin-Color and Template Based Technique for Automatic Ear Detection Surya Prakash, Umarani Jayaraman, Phalguni Gupta - Show less +2 more 04 Feb 2009 44 Citations	This paper proposes an efficient skin-color and template based technique for automatic ear detection in a side face image.
Open access•Journal Article•DOI Ear recognition based on Gabor features and KFDA. Li Yuan, Zhichun Mu - Show less +1 more 17 Mar 2014-The Scientific World Journal 21 Citations	We propose an ear recognition system based on 2D ear images which includes three stages: ear enrollment, feature extraction, and ear recognition.
Proceedings Article•DOI Fast and Fully Automatic Ear Detection Using Cascaded AdaBoost Syed Islam, Mohammed Bennamoun, Rowan Davies - Show less +2 more 07 Jan 2008 85 Citations	But accurate and rapid detection of the ear for real-time applications is a challenging task, particularly in the presence of occlusions.
Proceedings Article•DOI Human ear detection in the thermal infrared spectrum Ayman Abaza, Thirimachos Bourlai - Show less +1 more 18 May 2012-Proceedings of SPIE 12 Citations	Further analysis on our data set yielded that: (a) photometric normalization techniques do not directly improve ear detection performance.
Proceedings Article•DOI Ear authentication using Log-Gabor wavelets Ajay Kumar, Ajay Kumar, David Zhang - Show less +2 more 09 Apr 2007 32 Citations	Our experiments on two different public ear databases achieve promising results and suggest its utility in ear-based authentication.
Proceedings Article•DOI A comparative study on texture and surface descriptors for ear biometrics Anika Pflug, Pascal Nicklas Paul, Christoph Busch - Show less +2 more 18 Dec 2014 53 Citations	Recent research in texture-based ear recognition also indicates that ear detection and texture-based ear recognition are robust against signal degradation and encoding artefacts.
Open access•Journal Article•DOI 3D Ear Normalization and Recognition Based on Local Surface Variation Yi Zhang, Zhichun Mu, Li Yuan, Hui Zeng, Long Chen - Show less +4 more 21 Jan 2017-Applied Sciences 26 Citations	In this paper, an efficient and fully automatic 3D ear recognition system is proposed to address these issues.
Proceedings Article•DOI Ear detection based on improved AdaBoost algorithm Li Yuan, Feng Zhang - Show less +1 more 12 Jul 2009 31 Citations	The ear detection experiments on USTB ear database, CAS-PEAL face database and CMU PIE database show that the proposed method is significantly efficient and robust.

My columns

Related Questions

How do I turn off automatic ear detection on my Airpods?7 answers

How do I turn off crash detection on life360?8 answers

How do you turn off a smoke detector without going off?7 answers

How to connect noise Airpods to Android?18 answers

How to connect noise Airpods with Android?10 answers

How to pair noise Airpods with Android?6 answers

See what other people are reading

Are ensemble learning methods good with complex decision boundaries?

Ensemble learning methods excel in handling complex decision boundaries by combining multiple models to enhance predictive performance. Various approaches have been proposed to address this, such as modifying contrastive learning objectives to automatically train self-complementing ensembles, utilizing a novel voting model based on ensemble learning for threat detection in complex network environments, and designing a dynamic ensemble selection decision-making method to tackle class-imbalance and other data-complexity issues effectively. Additionally, a novel ensemble method called BEBS leverages ensemble models to capture robust decision boundaries in imbalanced data learning scenarios. Furthermore, an ensemble learning method based on deep neural networks and group decision making has been developed to generate diverse individual learners and combine them effectively for improved classification performance.How does the temperature changes in the smelting furnace on the quality of aluminum?

Temperature changes in the smelting furnace significantly impact the quality of aluminum. A temperature control system ensures precise temperature regulation, enhancing production stability and product quality. Predicting furnace box temperature through a deep belief network aids in maintaining stable temperatures, preventing damage to components, and maximizing economic benefits. Automated smelting furnaces provide safety for workers, uniform material distribution, and high-quality smelting due to consistent heating. The utility model of an aluminum alloy smelting furnace addresses the challenge of adjusting heater temperature based on molten aluminum and furnace atmosphere, optimizing energy consumption and quality control. A high-purity aluminum smelting furnace with integrated heating, cooling, and temperature monitoring devices improves efficiency and enhances the quality of purified aluminum.Customer sastification in ecommerce?

Customer satisfaction in e-commerce is crucial for business success. Various studies have focused on customer segmentation and satisfaction in the e-commerce domain. Machine learning algorithms like Support Vector Machine, Logistic Regression, K-Nearest Neighbors, Decision Tree, Random Forest, AdaBoost, and Gradient Boosting have been utilized for customer segmentation. These algorithms help in analyzing customer behaviors, forming clusters, and achieving high precision levels. Additionally, the use of unsupervised machine learning methods like the RFM model, TF-IDF method, K-means clustering, and Apriori Algorithm aids in systematic customer segmentation based on transaction data. Furthermore, text comments in e-commerce clothing reviews play a significant role in understanding customer sentiments and preferences, enabling the implementation of neural network models for forecasting customer reviews and predicting reviewer characteristics.How to use prior information to do image classification?

To utilize prior information for image classification, various methods have been proposed in research. One approach involves incorporating Bayesian methods to estimate prior probabilities of classes, enhancing classification accuracy by classifying ambiguous pixels based on remotely sensed imagery or existing land-cover maps. Another method introduces Semantic Based Regularization, integrating prior knowledge expressed as first-order logic clauses into deep learning models to boost accuracy in image classification tasks. Additionally, leveraging crowdsourced OpenStreetMap (OSM) data has shown promise in providing prior information for high-resolution image classification, refining training sets and superimposing road information to improve classification results. Furthermore, a novel classifier, the prior knowledge-based probabilistic collaborative representation-based classifier (PKPCRC), incorporates characteristics of training samples as prior knowledge to enhance visual recognition accuracy.What are the authors' proposed contour and region harmonic features for sub-local facial expression recognition?

The authors propose various methods for sub-local facial expression recognition. Zhong et al. suggest a framework based on local regions and key frames, utilizing dual-cross patterns for feature extraction. Qiu et al. introduce a Sliding Window Attention Network (SWA-Net) that includes a sliding window strategy for feature-level cropping and a local feature enhancement module for mining fine-grained features. Additionally, Feng and Shao present a local-face expression recognition model based on migration learning, utilizing Inception-v3 pre-training model and various image preprocessing techniques for accurate recognition. These approaches focus on enhancing feature extraction from specific facial regions to improve the accuracy of facial expression recognition.What is dependent on of selection of algorithm credit risk prediction?

The selection of the algorithm for credit risk prediction is crucial as it directly impacts the accuracy and efficiency of the prediction models. Various machine learning algorithms like XGBoost, CatBoost, Random Forest, and neural networks have been compared for credit risk prediction, with gradient boosting algorithms like XGBoost and CatBoost demonstrating superior performance in terms of training and testing accuracy. Additionally, the use of integrated learning methods, such as combining feature selection techniques like Relief algorithm and Random Forest with adaptive sampling methods like Borderline-SMOTE, significantly improves the prediction results by addressing high-dimensional features and data imbalance issues. Therefore, the careful selection of algorithms and integrated learning techniques plays a vital role in enhancing the effectiveness of credit risk prediction models.How to analyze the target detection results in automatic driving?

Analyzing target detection results in automatic driving involves a multifaceted approach that incorporates various methodologies and metrics to ensure the accuracy, speed, and reliability of detection systems. Firstly, the mean Average Precision (mAP) is a critical metric for evaluating the accuracy of target detection algorithms, as demonstrated by the use of the YOLOv5s network, which achieved a significant detection accuracy and speed in various weather conditions. The introduction of the DCN network and the use of K-means++ clustering algorithm for anchor point selection have shown improvements in mAP on standard datasets like KITTI and COCO, indicating a more suitable target detection for automatic driving scenarios. Moreover, the integration of advanced mechanisms such as the ACmix attention mechanism can further enhance the model's ability to focus on different target regions, thereby improving detection accuracy on datasets like BDD100K and PASCAL VOC2007. The use of connected automated vehicle (CAV) cooperative perception for data acquisition and processing, including object detection and tracking, provides a comprehensive analysis platform that leverages deep learning and sensor fusion. Additionally, the calculation of presence and coherence indicators can quantify the probability and predictability of target objects, respectively, offering a nuanced approach to analyzing detection results and adjusting driving instructions accordingly. The evaluation of autonomous vehicles through longitudinal relative distance computation and scenario-based testing further aids in the comparative analysis of detection accuracy under various conditions. The employment of expanded convolution modules and feature pyramid structures in road target identification methods addresses the challenge of detecting small targets by increasing the receptive field of network layers, which is crucial for analyzing detection results in complex road environments. The implementation of deep learning and binocular camera shooting enhances target detection and positioning accuracy, addressing the adaptability to complex road conditions. Lastly, understanding the current landscape of target detection algorithms and their challenges is essential for analyzing and improving autonomous driving systems.Is frequency following response a unique neural signature of all individuals?

The frequency-following response (FFR) serves as a distinctive neural marker that varies among individuals. It reflects phase-locked auditory encoding, capturing individual differences in auditory processing and specialized skills like absolute pitch labeling. FFR patterns emerge uniquely in infants over time, indicating individual biometric distinctions. Studies show that FFR strength declines from subcortical to cortical sources, with subcortical sources predominantly influencing speech perception skills. Moreover, FFR responses differ between individuals who can sing in-tune and those who sing off-tune, suggesting distinct neural processing patterns. Therefore, the FFR indeed represents a unique neural signature that varies across individuals, reflecting differences in auditory processing, skill specialization, and sensitivity to specific auditory stimuli.What is the current state of FIB SEM image analysis in the field of shale gas sample analysis?

The current state of FIB-SEM image analysis in shale gas sample analysis involves advancements in characterizing pore structures and connectivity. Researchers have utilized FIB-SEM for high-resolution imaging to distinguish organic and inorganic pores in shale samples. Automated workflows combining feature extraction and machine learning have been developed for accurate segmentation of SEM images, particularly for identifying rock components like pores, matrix, pyrite, and organic matter. Additionally, studies have shown that utilizing large volume FIB-SEM models significantly improves the representativeness of shale pore structures, enabling more accurate characterization for shale oil exploration and development. These advancements in FIB-SEM image analysis contribute to a better understanding of shale reservoir properties and aid in assessing the potential of gas shales for energy exploration.How are AI and ML methods used to characterize soil?

AI and ML methods are utilized in soil characterization through various approaches. Machine learning algorithms, including Random Forests, Support Vector Machine, Decision Trees, and eXtreme Gradient Boosting, are employed to predict soil shear wave velocity (Vs) from cone penetration test (CPT) data. Additionally, ML algorithms such as artificial neural network (ANN), Random Forest, Support Vector Machine, and Decision Trees are used to classify soils based on Robertson’s soil behavioral types, achieving high accuracy in soil classification. These methods offer efficient and accurate soil classification, reducing the need for extensive field and laboratory work. Furthermore, AI and ML models are developed to classify soil behavior types based on CPT results, with Random Forest and Extreme Gradient Boosting algorithms demonstrating superior performance in soil prediction and model training. Overall, AI and ML play a crucial role in streamlining soil characterization processes and enhancing accuracy in soil classification.Did not encounter unexpected or unusual noise from equipment malfunctions.fabric industry?

In the fabric industry, unexpected or unusual noise from equipment malfunctions can be a significant concern impacting productivity and product quality. Research has shown that noise caused by machines in textile industries is a critical issue, with studies focusing on measuring noise levels in dyeing and clothes preparation companies to assess worker exposure levels and recommend noise reduction measures. Additionally, a study on fault detection systems in textile machinery highlighted the importance of sound-based approaches using machine learning models trained on large datasets to detect machinery faults promptly and accurately, thus addressing limitations of existing methods. These findings emphasize the relevance of monitoring and addressing equipment malfunctions through sound-based anomaly detection systems in the fabric industry to ensure operational efficiency and product quality.