Search

Global Land Cover (LC) datasets are important geo-information sources for environmental, climate, agriculture, and landscape applications. The ESA WorldCover project (2020 and 2021) provides a global scale land cover map with the predefined 11 generic classes for almost the current state. This study aims to evaluate the accuracy of the ESA LC data for a local region in Lalapasa/Edirne to provide insights into this data for possible local-level applications. Our study revealed that while the grassland, shrubland, and bare classes have inaccuracies that need to be further addressed, tree cover, water bodies, and cropland LC classified were correctly mapped for the studied region. We proposed strategies to improve the accuracy of some classes in the ESA LC map with integrated usage of open geospatial datasets and object-based classification. We encourage merging segments with their best-fitting surrounding segments if they are smaller than a minimum mapping unit of 1 ha. By doing this, we aim to improve the representation of the integrity and compactness of built-up regions and agricultural lands., European Union (EU; Horizon 2020; Research and Innovation Program Grant; European Research Council (ERC); “A GeoAI-based Land Use Land Cover Segmentation Process to Analyse and Predict Rural Depopulation, Agricultural Land Abandonment, and Deforestation in Bulgaria and Turkey, 1940-2040” Project; GeoAI_LULC_Seg

This study aims to generate a new benchmark dataset from historical panchromatic aerial photographs suitable for deep learning-based Land use/Land cover (LULC) segmentation task. This new benchmark dataset spans a wide geographic area and consists of aerial photographs from various populous areas in Turkey and Bulgaria from the 1950s, 1960s, and 1970s. We implemented U-Net++ and Deeplabv3 segmentation architectures and appropriate hyperparameters and backbone structures to determine the applicability of this dataset, specifically for accurate and fast mapping of past terrain conditions. This unique historical LULC dataset and the different combinations of deep learning experiments proposed can be applied to different geographical regions with similar panchromatic datasets., European Union (EU; Horizon 2020; Research and Innovation Program Grant; European Research Council (ERC); “A GeoAI-based Land Use Land Cover Segmentation Process to Analyse and Predict Rural Depopulation, Agricultural Land Abandonment, and Deforestation in Bulgaria and Turkey, 1940-2040” Project; GeoAI_LULC_Seg; Istanbul Technical University, Scientific Research Unit (ITU-BAP)

Automatic road extraction from historical maps is an important task to understand past transportation conditions and conduct spatiotemporal analysis revealing information about historical events and human activities over the years. This research aimed to propose the ideal architecture, encoder, and hyperparameter settings for the historical road extraction task. We used a dataset including 7076 patches with the size of 256 x 256 pixels generated from scanned historical Deutsche Heereskarte 1:200 000 Turkei (DHK 200 Turkey) maps and their corresponding digitized ground truth masks for five different roads types. We first tested the widely used Unet++ and Deeplabv3 architectures. We also evaluated the contribution of attention models by implementing Unet++ with the concurrent spatial and channel-squeeze and excitation block and multiscale attention net. We achieved the best results with split-attention network (Timm-resnest200e) encoder and Unet++ architecture, with 98.99% overall accuracy, 41.99% intersection of union, 51.41% precision, 69.7% recall, and 57.72% F1 score values. Our output weights could be directly used for the inference of other DHK maps and transfer learning for similar or different historical maps. The proposed architecture could also be implemented in different road extraction studies., European Research Council (ERC); Project of “Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850-2000”; European Union (EU); Horizon 2020; Research and Innovation Program; UrbanOccupationsOETR

Agricultural land abandonment is a globally significant threat to the sustenance of economic, ecological, and social balance. Although the driving forces behind it can be multifold and versatile, rural depopulation and urbanization are significant contributors to agricultural land abandonment. In our chosen case study, focusing on two locations, Ruen and Stamboliyski, within the Plovdiv region of Bulgaria, we use aerial photographs and satellite imagery dating from the 1950s until 1980, in connection with official population census data, to assess the magnitude of agricultural abandonment for the first time from a remote sensing perspective. We use multi-modal data obtained from historical aerial and satellite images to accurately identify Land Use Land Cover changes. We suggest using the rubber sheeting method for the geometric correction of multi-modal data obtained from aerial photos and Key Hole missions. Our approach helps with precise sub-pixel alignment of related datasets. We implemented an iterative object-based classification approach to accurately map LULC distribution and quantify spatio-temporal changes from historical panchromatic images, which could be applied to similar images of different geographical regions., European Research Council (ERC); Project of Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850–2000; European Union (EU); Horizon 2020; Research and Innovation Program; Urban Occupations OETR

Deep learning-based segmentation of very high-resolution (VHR) satellite images is a significant task providing valuable information for various geospatial applications, specifically for land use/land cover (LULC) mapping. The segmentation task becomes more challenging with the increasing number and complexity of LULC classes. In this research, we generated a new benchmark dataset from VHR Worldview-3 images for twelve distinct LULC classes of two different geographical locations. We evaluated the performance of different segmentation architectures and encoders to find the best design to create highly accurate LULC maps. Our results showed that the DeepLabv3+ architecture with an ResNeXt50 encoder achieved the best performance for different metric values with an IoU of 89.46%, an F-1 score of 94.35%, a precision of 94.25%, and a recall of 94.49%. This design could be used by other researchers for LULC mapping of similar classes from different satellite images or for different geographical regions. Moreover, our benchmark dataset can be used as a reference for implementing new segmentation models via supervised, semi- or weakly-supervised deep learning models. In addition, our model results can be used for transfer learning and generalizability of different methodologies., European Union (EU); Horizon 2020; European Research Council (ERC); “Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850–2000” Project; Research and Innovation Program; UrbanOccupationsOETR

Agricultural land cover and its changing extent are directly related to human activities, which have an adverse impact on the environment and ecosystems. The historical knowledge of crop production and its cultivation area is a key element. Such data provide a base for monitoring and mapping spatio-temporal changes in agricultural land cover/use, which is of great significance to examine its impacts on environmental systems. Historical maps and related data obtained from historical archives can be effectively used for reconstruction purposes through using sample data from ground observations, government inventories, or other historical sources. This study considered historical population and cropland survey data obtained from Ottoman Archives and cropland suitability map, accessibility, and geophysical attributes as ancillary data to estimate non-irrigated crop production and its corresponding cultivation area in the 1840s Bursa Region, Turkey. We used the regression analysis approach to estimate agricultural land area and grain production for the unknown data points in the study region. We provide the spatial distribution of production and its cultivation area based on the estimates of regression models. The reconstruction can be used in line with future historical research aiming to model landscape, climate, and ecosystems to assess the impact of human activities on the environmental systems in preindustrial times in the Bursa Region context.

Although mountainous regions remained relatively isolated and almost untouched by the Ottoman rule, labor migration connected the inhabitants of these regions to the socioeconomic and political processes in the Ottoman Empire and beyond. Krusevo, a highland village located in present-day North Macedonia, provides an excellent case for understanding these connections. This paper presents systematic evidence from the Ottoman archives to document and analyze the social, economic, and demographic impacts of labor migration during this period. It provides an in-depth analysis of the Ottoman population and tax records of Krusevo in the 1840s, demonstrating the occupational profiles, migration patterns, and family and neighborhood networks of village residents during this period. Based on this analysis, it argues that labor migration was key to the transformation of social, economic, and demographic relations in rural communities and to the integration of even the most remote highland villages with the modernization processes that characterized the Ottoman Empire in the 19th century.

Scanned historical maps are available from different sources in various scales and contents. Automatic geographical feature extraction from these historical maps is an essential task to derive valuable spatial information on the characteristics and distribution of transportation infrastructures and settlements and to conduct quantitative and geometrical analysis. In this research, we used the Deutsche Heereskarte 1:200,000 Türkei (DHK 200 Turkey) maps as the base geoinformation source to construct the past transportation networks using the deep learning approach. Five different road types were digitized and labeled to be used as inputs for the proposed deep learning-based segmentation approach. We adapted U-Net++ and ResneXt50_32×4d architectures to produce multi-class segmentation masks and perform feature extraction to determine various road types accurately. We achieved remarkable results, with 98.73% overall accuracy, 41.99% intersection of union, and 46.61% F1 score values. The proposed method can be implemented in DHK maps of different countries to automatically extract different road types and used for transfer learning of different historical maps.

The historic reconstruction of residential land cover is of significance to uncover the hu-man-environment relationship and its changing dynamics. Taking into account the historical census data and cadastral maps of seven villages, this study generated residential land cover maps for the Bursa Region in the 1850s using a model based on natural constraints, land zoning, socio-economic factors and residential suitability. Two different historical reconstructions were generated; one based on a high density residential model and another based on a low density model. The simulated landcover information was used as an ancillary data to redistribute aggregated census counts to fine scale raster cells. Two different statistical models were developed; one based on probability maps and the other applying regression models including Ordinary Least Squares (OLS) and Geographically Weighted Regression (GWR) models. The regression models were validated with historical census data of the 1840s. From regression models, socio-economic and physical characteristics, accessibility and natural amenities showed significant impacts on the distribution of population. Model validation analysis revealed that GWR is more accurate than OLS models. The generated residential land cover and gridded population datasets can provide a basis for the historical study of population and land use., European Union (EU); Horizon 2020; European Research Council (ERC); Research and Innovation Programme; Project: ""Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850-2000""

Scanned historical maps are available from different sources in various scales and contents. Automatic geographical feature extraction from these historical maps is an essential task to derive valuable spatial information on the characteristics and distribution of transportation infrastructures and settlements and to conduct quantitative and geometrical analysis. In this research, we used the Deutsche Heereskarte 1:200,000 Türkei (DHK 200 Turkey) maps as the base geoinformation source to construct the past transportation networks using the deep learning approach. Five different road types were digitized and labeled to be used as inputs for the proposed deep learning-based segmentation approach. We adapted U-Net++ and ResneXt50_32×4d architectures to produce multi-class segmentation masks and perform feature extraction to determine various road types accurately. We achieved remarkable results, with 98.73% overall accuracy, 41.99% intersection of union, and 46.61% F1 score values. The proposed method can be implemented in DHK maps of different countries to automatically extract different road types and used for transfer learning of different historical maps., European Union (EU); Horizon 2020; European Research Council (ERC); Research and Innovation Program; Project “Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850–2000”; UrbanOccupationsOETR

With the increased amount of digitized historical documents, information extraction from them gains pace. Historical maps contain valuable information about historical, geographical and economic aspects of an era. Retrieving information from historical maps is more challenging than processing modern maps due to lower image quality, degradation of documents and the massive amount of non-annotated digital map archives. Convolutional Neural Networks (CNN) solved many image processing challenges with great success, but they require a vast amount of annotated data. For historical maps, this means an unprecedented scale of manual data entry and annotation. In this study, we first manually annotated the Third Military Mapping Survey of Austria-Hungary historical map series conducted between 1884 and 1918 and made them publicly accessible. We recognized different road types and their pixel-wise positions automatically by using a CNN architecture and achieved promising results., European Union (EU); Horizon 2020; European Research Council (ERC);Research and innovation Programme; Project: ""Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850-2000""; UrbanOccupationsOETR

Recently, an increasing number of studies have applied deep learning algorithms for extracting information from handwritten historical documents. In order to accomplish that, documents must be divided into smaller parts. Page and line segmentation are vital stages in the Handwritten Text Recognition systems; it directly affects the character segmentation stage, which in turn deter-mines the recognition success. In this study, we first applied deep learning-based layout analysis techniques to detect individuals in the first Ottoman population register series collected between the 1840s and the 1860s. Then, we employed horizontal projection profile-based line segmentation to the demographic information of these detected individuals in these registers. We further trained a CNN model to recognize automatically detected ages of individuals and estimated age distributions of people from these historical documents. Extracting age information from these historical registers is significant because it has enormous potential to revolutionize historical demography of around 20 successor states of the Ottoman Empire or countries of today. We achieved approximately 60% digit accuracy for recognizing the numbers in these registers and estimated the age distribution with Root Mean Square Error 23.61., European Union (EU); Horizon 2020; European Research Council (ERC); Research and Innovation Program; “Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850–2000” Project; UrbanOccupationsOETR

Although mountainous regions remained relatively isolated and almost untouched by the Ottoman rule, labor migration connected the inhabitants of these regions to the socioeconomic and political processes in the Ottoman Empire and beyond. Krusevo, a highland village located in present-day North Macedonia, provides an excellent case for understanding these connections. This paper presents systematic evidence from the Ottoman archives to document and analyze the social, economic, and demographic impacts of labor migration during this period. It provides an in-depth analysis of the Ottoman population and tax records of Krusevo in the 1840s, demonstrating the occupational profiles, migration patterns, and family and neighborhood networks of village residents during this period. Based on this analysis, it argues that labor migration was key to the transformation of social, economic, and demographic relations in rural communities and to the integration of even the most remote highland villages with the modernization processes that characterized the Ottoman Empire in the 19th century., European Union (EU); Horizon 2020; European Research Council (ERC); Research and Innovation Programme; Project: ""Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850-2000""; UrbanOccupationsOETR

Agricultural land cover and its changing extent are directly related to human activities, which have an adverse impact on the environment and ecosystems. The historical knowledge of crop production and its cultivation area is a key element. Such data provide a base for monitoring and mapping spatio-temporal changes in agricultural land cover/use, which is of great significance to examine its impacts on environmental systems. Historical maps and related data obtained from historical archives can be effectively used for reconstruction purposes through using sample data from ground observations, government inventories, or other historical sources. This study considered historical population and cropland survey data obtained from Ottoman Archives and cropland suitability map, accessibility, and geophysical attributes as ancillary data to estimate non-irrigated crop production and its corresponding cultivation area in the 1840s Bursa Region, Turkey. We used the regression analysis approach to estimate agricultural land area and grain production for the unknown data points in the study region. We provide the spatial distribution of production and its cultivation area based on the estimates of regression models. The reconstruction can be used in line with future historical research aiming to model landscape, climate, and ecosystems to assess the impact of human activities on the environmental systems in preindustrial times in the Bursa Region context., European Union (EU); Horizon 2020; European Research Council (ERC); Research and innovation Programme; Project: ""Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850-2000""; UrbanOccupationsOETR

For extracting information from the historical documents, digitization efforts have increased dramatically in the recent decades. Accurate layout analysis will help researchers for developing more robust HTR and OCR techniques which will extract meaningful information from these documents. Variable layouts, low quality and distorted images of historical documents create different problems to deal with when compared to modern document processing. Arabic script features have even more problems for these automatic processing systems. In this study, we have developed a tool for automatically analyzing the layouts of the first Ottoman population registers which are written in Arabic script form. We built a dataset for testing the performance of our system which are chosen from the first population records of the Ottoman Empire between the 1840s and 1860s. We successfully classified two different object types in those documents. / Tarihsel belgelerden bilgi çıkarmak için sayısallaştırma çabaları son yıllarda önemli ölçüde artmı ştır. Doğru yerleşim analizi, araştırmacıların bu belgelerden anlamlı bilgiler elde edecek daha sağlam HTR ve OCR teknikleri geliştirmelerine yardımcı olacaktır. Değişken yerleşimler, düşük kaliteli ve bozuk tarihi belgeler, modern belge işlemeye kıyasla farklı sorunlar yaratmaktadır. Arapça yazılar kendine has bazı özelliklerinden dolayı otomatik işlem sistemleri için daha da fazla sorun yaratmaktadır. Bu çalışmada, ilk Osmanlı nüfus kayıtlarının Arap harfleriyle yazılmış yerleşimlerini otomatik olarak analiz etmek için bir araç geliştirdik. 1840’lar ve 1860’lar arasında Osmanlı İmparatorluğu’nun ilk nüfus kayıtlarından seçilen sistemimizin performansını test etmek için bir veri seti oluşturduk. Bu belgelerde iki farklı nesne türünü başarıyla sınıflandırdık., European Union (EU); Horizon 2020; European Research Council (ERC); Research and innovation Programme; Project: "Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850-2000"

For 25 years the Corona reconnaissance satellite mission has been declassified, making available an unprecedented historic archive of very high spatial resolution panchromatic images acquired during the cold war era. During the same time, Bulgaria has observed a decreasing and ageing population and a slow agricultural productivity. In this study, we attempt to map the settlements around the city of Plovdiv, Bulgaria based on textural analysis of a Corona image acquired in 1968. We compare the results with information derived from recent Sentinel-2B and Landsat-8 images. We discuss on the discerning capability of the Corona image in mapping settlements and the potential usability of textural analysis in the context of land use and land cover mapping of historical images. Overall, we present the textural analysis of an image from one of the first satellite missions and report on the potential for feature extraction from such primitive satellite products., European Union (EU); Horizon 2020; European Research Council (ERC); Research and innovation Programme; Project: "Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850-2000"; UrbanOccupationsOETR

European Union (EU); Horizon 2020; European Research Council (ERC); Research and innovation Programme; Project: "Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850-2000"; UrbanOccupationsOETR

The digitalization of historical documents continues to gain pace for further processing and extract meanings from these documents. Page segmentation and layout analysis are crucial for historical document analysis systems. Errors in these steps will create difficulties in the information retrieval processes. Degradation of documents, digitization errors and varying layout styles complicate the segmentation of historical documents. The properties of Arabic scripts such as connected letters, ligatures, diacritics and different writing styles make it even more challenging to process Arabic historical documents. In this study, we developed an automatic system for counting registered individuals and assigning them to populated places by using a CNN-based architecture. To evaluate the performance of our system, we created a labeled dataset of registers obtained from the first wave of population registers of the Ottoman Empire held between the 1840s–1860s. We achieved promising results for classifying different types of objects and counting the individuals and assigning them to populated places.

The purpose of this article is to announce the publication of a partial transcription of the c. 1907 Ottoman census that relates to 24,000 Apostolic Armenian Istanbulites and of a historical geographic information system (HGIS), or interactive map, that shows where these individuals lived. Within this framework, the authors first present their argument that an unidentified, microfilmed population register housed in New York is the most substantial portion of this census currently available to researchers. In the second part of their article, they introduce HGISes as tools for the digital humanities and describe the process of creating one., Calouste Gulbenkian Foundation

The Corona spy programme was a series of reconnaissance satellites which delivered more than 860000 images between 1960 and 1972. Since 1995, the data are declassified and a large historic earth observation archive is made accessible to the scientific community. Despite the large volume of information and the high spatial resolution of the satellite imagery, little has been done in the last 25 years in the context of image processing of this data source, a fact which perhaps can be attributed to the technical difficulties of these primitive images such as the lack of metadata, intense spatial and radiometric distortions, low Signal-to-Noise Ratio (SNR) and a single panchromatic band. Hence, the photogrammetric challenges to extract useful information are paramount. In this study, we present recent developments arising from our efforts to map settlements and agricultural parcels over the Plovdiv region, Bulgaria from a Corona image acquired in 1968. We, overall, present initial findings from the integration of earth observation into the ERC-StG project UrbanOccupationsOETR and evaluate the usability of such primitive images in feature extraction. We compare the areas corresponding to settlements and correlate them with concurrent population census. Based on the findings, we suggest that settlements and agricultural parcels can be mapped from a Corona KH-4B image with fine radiometric quality., European Union (European Union); European Research Council (ERC); Horizon 2020 Framework Programme

Historical manuscripts and archival documentation are handwritten texts which are the backbone sources for historical inquiry. Recent developments in the digital humanities field and the need for extracting information from the historical documents have fastened the digitization processes. Cutting edge machine learning methods are applied to extract meaning from these documents. Page segmentation (layout analysis), keyword, number and symbol spotting, handwritten text recognition algorithms are tested on historical documents. For most of the languages, these techniques are widely studied and high performance techniques are developed. However, the properties of Arabic scripts (i.e., diacritics, varying script styles, diacritics, and ligatures) create additional problems for these algorithms and, therefore, the number of research is limited. In this research, we first automatically spotted the Arabic numerals from the very first series of population registers of the Ottoman Empire conducted in the mid-nineteenth century and recognized these numbers. They are important because they held information about the number of households, registered individuals and ages of individuals. We applied a red color filter to separate numerals from the document by taking advantage of the structure of the studied registers (numerals are written in red). We first used a CNN-based segmentation method for spotting these numerals. In the second part, we annotated a local Arabic handwritten digit dataset from the spotted numerals by selecting uni-digit ones and tested the Deep Transfer Learning method from large open Arabic handwritten digit datasets for digit recognition. We achieved promising results for recognizing digits in these historical documents., European Research Council (ERC) Project: "Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850-2000" (; UrbanOccupationsOETR); European Union (European Union); Horizon 2020

The digitalization of historical documents continues to gain pace for further processing and extract meanings from these documents. Page segmentation and layout analysis are crucial for historical document analysis systems. Errors in these steps will create difficulties in the information retrieval processes. Degradation of documents, digitization errors and varying layout styles complicate the segmentation of historical documents. The properties of Arabic scripts such as connected letters, ligatures, diacritics and different writing styles make it even more challenging to process Arabic historical documents. In this study, we developed an automatic system for counting registered individuals and assigning them to populated places by using a CNN-based architecture. To evaluate the performance of our system, we created a labeled dataset of registers obtained from the first wave of population registers of the Ottoman Empire held between the 1840s–1860s. We achieved promising results for classifying different types of objects and counting the individuals and assigning them to populated places., European Union (EU); Horizon 2020; European Research Council (ERC); Research and innovation Programme; Project: "Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850-2000"

European Union (EU); Horizon 2020; European Research Council (ERC); Research and innovation Programme; Project: "Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850-2000"; UrbanOccupationsOETR

This research introduces a novel geo-spatial sampling model to overcome a major difficulty in historical economic geography of Bulgarian lands during a crucial period: immediately before and after the de facto independence of the territory from the Ottoman Empire in the second half of the nineteenth century. At its core it seeks to investigate the research question how the Bulgarian independence affected agricultural production in two regions (centered around the cities of Plovdiv and Ruse) of today's Bulgaria, for which there are conflicting yet empirically unsubstantiated claims concerning the economic impact of the political independence. Using our be-spoke geo-sampling strategy we believe, we have sampled regionally representative commensurable agricultural data from the 1840s Ottoman archival documentation, in accord with agricultural censuses conducted by the nascent nation state of Bulgaria in the 1890s., European Union (EU); Horizon 2020; European Research Council (ERC); Research and innovation Programme; Project: "Industrialisation and Urban Growth from the mid-nineteenth century Ottoman Empire to Contemporary Turkey in a Comparative Perspective, 1850-2000"; UrbanOccupationsOETR

Historical document analysis systems gain importance with the increasing efforts in the digitalization of archives. Page segmentation and layout analysis are crucial steps for such systems. Errors in these steps will affect the outcome of handwritten text recognition and Optical Character Recognition (OCR) methods, which increase the importance of the page segmentation and layout analysis. Degradation of documents, digitization errors, and varying layout styles are the issues that complicate the segmentation of historical documents. The properties of Arabic scripts such as connected letters, ligatures, diacritics, and different writing styles make it even more challenging to process Arabic script historical documents. In this study, we developed an automatic system for counting registered individuals and assigning them to populated places by using a CNN-based architecture. To evaluate the performance of our system, we created a labeled dataset of registers obtained from the first wave of population registers of the Ottoman Empire held between the 1840s and 1860s. We achieved promising results for classifying different types of objects and counting the individuals and assigning them to populated places., European Union (European Union); Horizon 2020; European Research Council (ERC); Research and Innovation Program Grant; UrbanOccupationsOETR