Classification of engraved pottery sherds mixing deep-learning features by compact bilinear pooling

Chetouani, Aladine; Treuillet, Sylvie; Exbrayat, Matthieu; Jesset, Sébastien

doi:10.1016/j.patrec.2019.12.009

Cited by 31 publications

(17 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, the ML analysis of chemical data for provenience studies that rely on cluster and factor analysis can be less influenced by the statistical requirements of those algorithms and can be refined as new data becomes available (Hazenfratz Marks et al 2017). Similarly, ML has been used for pattern classification of pottery styles (Bickler 2018a; Chetouani et al 2020; Romanengo et al 2020).…”

Section: Machine Learning For Archaeological Datamentioning

confidence: 99%

Machine Learning Arrives in Archaeology

Bickler¹

2021

Adv. archaeol. pract.

View full text Add to dashboard Cite

OverviewMachine learning (ML) is rapidly being adopted by archaeologists interested in analyzing a range of geospatial, material cultural, textual, natural, and artistic data. The algorithms are particularly suited toward rapid identification and classification of archaeological features and objects. The results of these new studies include identification of many new sites around the world and improved classification of large archaeological datasets. ML fits well with more traditional methods used in archaeological analysis, and it remains subject to both the benefits and difficulties of those approaches. Small datasets associated with archaeological work make ML vulnerable to hidden complexity, systemic bias, and high validation costs if not managed appropriately. ML's scalability, flexibility, and rapid development, however, make it an essential part of twenty-first-century archaeological practice. This review briefly describes what ML is, how it is being used in archaeology today, and where it might be used in the future for archaeological purposes.

show abstract

Section: Machine Learning For Archaeological Datamentioning

confidence: 99%

Machine Learning Arrives in Archaeology

Bickler¹

2021

Adv. archaeol. pract.

View full text Add to dashboard Cite

show abstract

“…Some researchers proposed their own models [45] trained from scratch, while others employed pre-trained models like AlexNet [72] and ResNet [81]. In our technique, we used the model introduced by the Oxford Visual Geometry Group (VGG), as this model is widely utilized and provided decent results in many applications [67,[82][83][84][85][86]. More precisely, we fine-tuned the pre-trained VGG16 model without data augmentation, since these treatments change the structure of the data and thus modify the perceived quality [87].…”

Section: Cnn Modelmentioning

confidence: 99%

Image Quality Assessment without Reference by Combining Deep Learning-Based Features and Viewing Distance

Chetouani

Pedersen

2021

Applied Sciences

Self Cite

View full text Add to dashboard Cite

An abundance of objective image quality metrics have been introduced in the literature. One important essential aspect that perceived image quality is dependent on is the viewing distance from the observer to the image. We introduce in this study a novel image quality metric able to estimate the quality of a given image without reference for different viewing distances between the image and the observer. We first select relevant patches from the image using saliency information. For each patch, a feature vector is extracted from a convolutional neural network model and concatenated at the viewing distance, for which the quality is predicted. The resulting vector is fed to fully connected layers to predict subjective scores for the considered viewing distance. The proposed method was evaluated using the Colourlab Image Database: Image Quality and Viewing Distance-changed Image Database. Both databases provide subjective scores at two different viewing distances. In the Colourlab Image Database: Image Quality we obtain a Pearson correlation of 0.87 at both 50 cm and 100 cm viewing distances, while in the Viewing Distance-changed Image Database we obtained a Pearson correlation of 0.93 and 0.94 at viewing distance of four and six times the image height. The results show the efficiency of our method and its generalization ability.

show abstract

“…In (6), the color similarity is normalized to a value between 0 and 100 for ease of calculation. (6) In this paper, we use the similarity measure as shown in (7). In 7 (7) In this paper, we define the similarity measure like (8) to calculate the final similarity between a target image and a test image by integrating similarity measure of color feature and texture feature.…”

Section: Detection Of Faulty Imagesmentioning

confidence: 99%

“…(6) In this paper, we use the similarity measure as shown in (7). In 7 (7) In this paper, we define the similarity measure like (8) to calculate the final similarity between a target image and a test image by integrating similarity measure of color feature and texture feature. In (8),  and  are the weighting factors [26] that represent the contribution of color and texture features.…”

Section: Detection Of Faulty Imagesmentioning

confidence: 99%

“…In particular, these devices have high-speed wired and wireless network functions and location-based sensors, thus the functions of the applications being developed are becoming more diverse [1][2][3][4][5]. Applications developed in this way run on a variety of platforms, and they are actually applied and used in various fields related to the Fourth Industrial Revolution, including artificial intelligence, autonomous driving, robots, drones, 3D printers, virtual reality and extended reality [6][7][8][9][10].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Effective Image Quality Inspection Based on Color and Texture Features

Jang*,

Lee*

2020

IJRTE

View full text Add to dashboard Cite

As people’s interest in software increases, the number and types of newly developed applications increase day by day. Therefore, there are attempts to automatically check whether the newly developed interface of these programs works correctly through image processing instead of manually. In this paper, we describe a strategy to robustly detect execution error screens of applications through comparative analysis of key features from various kinds of image scenes. In the approach described in this study, we extract the main multiple features representing the image. Then, by comparing the difference of the extracted multiple features, it is effectively determined whether the image is the same normal image as the reference image or an error image similar to the target image but different from each other. Experimental results verify that the described algorithm accurately detects normal and faulty images by comparing the main multiple features from various types of images. For performance evaluation, we quantitatively compared the performance of the introduced dissimilar region extraction strategy with those of other conventional methods. The existing methods of extracting similarity between images based on one feature or histogram comparison mostly lack information to sufficiently express the entire image. Therefore, when similar areas between two images are extracted, many errors occurred due to incorrect matching. However, the proposed multiple feature-based method obtains richer information by extracting multiple features from the input image. In addition, because the matching is done based on this information, the accuracy is relatively high. The approach introduced in this study is expected to be effectively used in many practical applications related to image processing such as image retrieval, trademark comparison, video security, application quality inspection, etc.

show abstract

Classification of engraved pottery sherds mixing deep-learning features by compact bilinear pooling

Cited by 31 publications

References 24 publications

Machine Learning Arrives in Archaeology

Machine Learning Arrives in Archaeology

Image Quality Assessment without Reference by Combining Deep Learning-Based Features and Viewing Distance

Effective Image Quality Inspection Based on Color and Texture Features

Contact Info

Product

Resources

About