Explore the vast range of titles available.

The severe drought of 1983–1984 in the Sahel region, and its socio‐economic impacts for people relying on farming had for consequence the first major gold rush at Koma Bangou in the southwestern part of Niger. Initiated in 1984, the gold panning activities were interrupted from 1989 to 1999 with exploration permits assigned to the mining industry. The site was reclassified at the year‐end 1999 as a gold panning site and artisanal mining resumed until present‐day. Gold panning activities such as ore extraction and cyanide processing produced mining waste including rocks, mine tailings, and treatment residues. Mining waste is a serious environmental, health and safety problem. Multispectral Landsat images (TM4‐5, ETM7+, OLI/TIRS) acquired between 1984 and 2020 were used to map the spatial evolution of waste generated by gold panning activities at Koma Bangou. Different processing methods were tested, including Minimum Noise Fraction (MNF) transform, Band Ratio (BR), and Feature Oriented Principal Component Selection (FPCS). The FPCS applied to hydroxyl‐bearing minerals appears to be most efficient to map gold extraction and cyanidation waste areas. The waste surface associated with ore extraction has increased from 9.43 ha in 1984 to 234.20 ha in 2020, with continuous expansion during the period of clandestine activity (1989–1999). The waste surface associated with cyanidation has increased from 5.56 ha in 2009 (the year of cyanide treatment introduction) to 99.53 ha in 2020. Landsat multispectral imagery proved a suitable data source for monitoring the evolution of gold mining waste and consequences of public policies at Koma Bangou. Plain Language Summary Artisanal gold mining in Niger is an alternative income‐generating activity for farmers. The first gold rush took place in Koma Bangou, in 1984, following the drought of 1983–1984 in order to cope with famine. The gold panning started in 1984, was interrupted during 1989–1999, and then resumed until present day. This work is based on multispectral analysis of almost 40 years of Landsat data to evaluate the evolution of the surface areas associated with gold extraction and cyanidation (cyanidation is a hydrometallurgical technique for extracting gold from ore by converting the gold to a water‐soluble coordination complex). Several image processing methods were tested for the purpose of mapping the extension of waste surfaces from 1984 onwards and the area of cyanidation waste from the year of introduction of this technique in 2009. Using the most efficient method, we report that the waste surface associated with gold extraction has increased from 9.43 ha in 1984 to 234.20 ha in 2020 with continuous expansion during the period of clandestine activity (1989–1999). The cyanidation areas have increased from 5.56 ha in 2009 to 99.53 ha in 2020. Key Points Koma Bangou is the major artisanal gold mining site in Niger and one of the major sites of artisanal gold mining in the Sahel Gold panning in Koma Bangou produces waste that degrades and pollutes the environment The use of Landsat satellite sensors allows to monitor the evolution of mining activities during four decades (1984–2020)

Recognition of degraded printed compound Kannada characters is a challenging research problem. It has been verified experimentally that noise removal is an essential preprocessing step. Proposed are two methods for degraded Kannada character recognition problem. Method 1 is conventionally used histogram of oriented gradients (HOG) feature extraction for character recognition problem. Extracted features are transformed and reduced using principal component analysis (PCA) and classification performed. Various classifiers are experimented with. Simple compound character classification is satisfactory (more than 98% accuracy) with this method. However, the method does not perform well on other two compound types. Method 2 is deep convolutional neural networks (CNN) model for classification. This outperforms HOG features and classification. The highest classification accuracy is found as 98.8% for simple compound character classification. The performance of deep CNN is far better for other two compound types. Deep CNN turns out to better for pooled character classes.

Handwritten signatures have traditionally been used as a common form of recognition and authentication in tasks such as financial transactions and document authentication. However, there are few studies on minority languages such as Uyghur and Kazakh used in Xinjiang, China, and no available public dataset for these scripts, which are widely used in banking and other fields. Therefore, this paper addresses this problem by constructing a dataset containing Uyghur, Kazakh, and Han languages and presents an automatic handwritten signature recognition approach based on Uyghur, Kazakh, Han, and public datasets. In the paper, a handwritten signature recognition method that combines local maximum occurrence features (LOMO) and histogram of orientated gradients (HOG) features was proposed. LOMO features use a sliding window to represent the local features of the signature image. The high-dimensional features formed by the combination of these methods are dimensionally reduced by principal component analysis (PCA). The classification is performed using k-nearest neighbors (k-NN), and it is compared with the random forest method. The proposed method achieved a recognition rate of 98.4% using a diverse signature database compared with existing methods. It shows that the method was effective and can be applied to large datasets of mixed, multilingual signatures.

This paper proposes a method of reducing dimensions based on genetic algorithm and object-oriented classification based on support vector machine (SVM). The basic idea is subspace decomposition of hyperspectral images at first, then selecting suitable bands in each subspace by using genetic algorithm and putting all selected bands of each subspace together. Furthermore, the hyperspectral image is segmented into a series of objects and then the spectral features and spatial features of objects in the selected bands are extracted. Finally, SVM classification is used according to features of the objects. The algorithm proposed is more effective and superior in dimension reduction and classification of hyperspectral image. [PUBLICATION ABSTRACT]