A Deep Convolutional Neural Network for Food Detection and Recognition

Subhi, Mohammed Ahmed; Ali, Sawal Hamid Md

doi:10.1109/iecbes.2018.8626720

Cited by 45 publications

(18 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Liu et al [37] explored deep learning for food classification and achieved a competitive accuracy. Myers et al [38], Zhang et al [39], Subhi and Ali [40], Yanai and Kawano [41], Christodoulidis et al [42] Wu et al [43] focused on different variations of deep learning for classification of food data sets. Some of these studies generated their own food datasets while other studies used already available food benchmarks for evaluation of their proposed approaches.…”

Section: Convolutional Neural Networkmentioning

confidence: 99%

An Open-Ended Continual Learning for Food Recognition Using Class Incremental Extreme Learning Machines

Tahir

Loo

2020

IEEE Access

View full text Add to dashboard Cite

State-of-the-art deep learning models for food recognition do not allow data incremental learning and often suffer from catastrophic interference problems during the class incremental learning. This is an important issue in food recognition since real-world food datasets are open-ended and dynamic, involving a continuous increase in food samples and food classes. Model retraining is often carried out to cope with the dynamic nature of the data, but this demands high-end computational resources and significant time. This paper proposes a new open-ended continual learning framework by employing transfer learning on deep models for feature extraction, Relief F for feature selection, and a novel adaptive reduced class incremental kernel extreme learning machine (ARCIKELM) for classification. Transfer learning is beneficial due to the high generalization ability of deep learning features. Relief F reduces computational complexity by ranking and selecting the extracted features. The novel ARCIKELM classifier dynamically adjusts network architecture to reduce catastrophic forgetting. It addresses domain adaptation problems when new samples of the existing class arrive. To conduct comprehensive experiments, we evaluated the model against four standard food benchmarks and a recently collected Pakistani food dataset. Experimental results show that the proposed framework learns new classes incrementally with less catastrophic inference and adapts domain changes while having competitive classification performance. INDEX TERMS Food recognition, deep learning, open-ended continual learning, class incremental extreme learning machine, adaptive class incremental extreme learning machine, adaptive reduced class incremental kernel extreme learning machine.

show abstract

Section: Convolutional Neural Networkmentioning

confidence: 99%

An Open-Ended Continual Learning for Food Recognition Using Class Incremental Extreme Learning Machines

Tahir

Loo

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Depth-wise separable convolution by itself can save more than 80 to 90 percent of computation on 3x3 convolution layer as we can see in (12). By trying to calculate the differences using (12) with DK = 3 while depth-wise convolution will only take 1/N worth of computational power than the normal convolution.…”

Section: Depthwise Separable Convolution Layermentioning

confidence: 99%

“…Deep Convolutional Neural Network (DCNN) has been implemented and set a breakthrough in many fields. Some of the fields are image recognition [1], segmentation [2][3][4][5][6][7], super-resolution [8,9], and many others [10][11][12]. The current state-of-the-arts are including AlexNet [13], ResNet [14], MobileNet [15], and other DCNN which based on Convolutional Neural Network (CNN) [16][17][18][19][20][21][22][23][24][25][26].…”

Section: Introductionmentioning

confidence: 99%

Small and Slim Deep Convolutional Neural Network for Mobile Device

2020

View full text Add to dashboard Cite

Recent development of deep convolutional neural networks (DCNN) devoted in creating a slim model for devices with lower specification such as embedded, mobile hardware, or microcomputer. Slim model can be achieved by minimizing computational complexity which theoretically will make processing time faster. Therefore, our focus is to build an architecture with minimum floating-point operation per second (FLOPs). In this work, we propose a small and slim architecture which later will be compared to state-of-the-art models. This architecture will be implemented into two models which are CustomNet and CustomNet2. Each of these models implements 3 convolutional blocks which reduce the computational complexity while maintains its accuracy and able to compete with state-of-the-art DCNN models. These models will be trained using ImageNet, CIFAR 10, CIFAR 100 and other datasets. The result will be compared based on accuracy, complexity, size, processing time, and trainable parameter. From the result, we found that one of our models which is CustomNet2, is better than MobileNet, MobileNet-v2, DenseNet, NASNetMobile in accuracy, trainable parameter, and complexity. For future implementation, this architecture can be adapted using region based DCNN for multiple object detection.

show abstract

“…GrabCut is an interactive segmentation algorithm that uses GMM (Gaussian mixture model) to estimate the colour distribution of the segmented object and background based on the specified bounding box of the segmented object [12] [13]. Equation (1) shows the energy function of the GrabCut.…”

Section: The Theoretical Modelmentioning

confidence: 99%

“…With the improvement of living standards, people began to pursue a more scientific and healthy diet. The food image classification is to automatically analyze the food images provided by the user through a computer and give a matching food name to further predict the user's diet and nutrient intake [1].…”

Section: Introductionmentioning

confidence: 99%

An SLGC Model for Asian Food Image Classification

Wu¹,

Zhao²,

Qu³

2020

JCC

View full text Add to dashboard Cite

As a fine-grained classification problem, food image classification faces many difficulties in the specific implementation. Different countries and regions have different eating habits. In particular, Asian food images have a complicated structure, and the related classification methods are still very scarce. There is an urgent need to develop a feature extraction and fusion scheme based on the characteristics of Asian food images. To solve the above problems, we proposed an image classification model SLGC (SURF-Local and Global Color) that combines image segmentation and feature fusion. By studying the unique structure of Asian foods, the color features of the images are merged into the representation vectors in the local and global dimensions, respectively, thereby further enhancing the effect of feature extraction. The experimental results show that the SLGC model can express the intrinsic characteristics of Asian food images more comprehensively and improve classification accuracy.

show abstract

A Deep Convolutional Neural Network for Food Detection and Recognition

Cited by 45 publications

References 8 publications

An Open-Ended Continual Learning for Food Recognition Using Class Incremental Extreme Learning Machines

An Open-Ended Continual Learning for Food Recognition Using Class Incremental Extreme Learning Machines

Small and Slim Deep Convolutional Neural Network for Mobile Device

An SLGC Model for Asian Food Image Classification

Contact Info

Product

Resources

About