Real Landmines Research Articles

Determining the effectiveness of using three-dimensional printing to train computer vision systems for landmine detection

The object of this study is the effectiveness of using three-dimensional printing to train computer vision models for landmine detection. The ongoing war in Ukraine has resulted in significant landmine contamination, particularly after russia’s full-scale invasion in 2022. Given the enormous amount of potentially landmine-contaminated land, fast and efficient demining techniques are required, as human probing and metal detectors are labor-intensive and slow-moving. Machine learning offers promising solutions to speed up the landmine detection process by deploying recognition models on robots and unmanned aerial vehicles. However, training such systems faces certain challenges. Firstly, the number of annotated data available for training is limited, which can hinder the model’s ability to generalize to real-world scenarios. Secondly, the use of real or even defused landmines is dangerous due to the potential for accidental detonation. This study aims to overcome the problem of limited data and the risk of using real landmines. Three-dimensional printing makes it possible to create safe and diverse training data, which is essential for model performance. The model trained on replicas, achieved 98 % and 91 % precision on printed and actual landmines, respectively. This high precision is attributed to the realism of copies and the use of advanced machine learning algorithms. This approach successfully addressed the research problem due to the safety, accessibility and diversity of copies. The models trained on copies of landmines could be used in humanitarian demining operations. These operations often employ unmanned aerial vehicles or robots to identify landmines that are thrown remotely, exposed on the surface, or partially hidden

Review of strategies to overcome the lack of data in landmine detection

This article reviews strategies to address the lack of data for training landmine recognition models. Since the outbreak of the war in Ukraine in 2014, the area of contaminated territories has gradually increased. However, after Russia's full-scale invasion on February 24, 2022, the problem of landmines in Ukraine has become much more severe, as the area of mined territories in the country has increased to 30%. It takes many years and efforts to clear such a large area. To overcome a problem of this scale, it is necessary to look for new methods of landmine detection that will allow demining to be conducted 24/7. Machine learning is one of the options for solving the problem of landmine detection. To train landmine recognition models, a large amount of data is required. However, the lack of diverse and large datasets creates significant obstacles to the development of effective detection systems. The safety concerns associated with conducting experiments with real landmines further exacerbate the problem. This article discusses three possible strategies to overcome the above problem: augmentation methods, the use of 3D printing technology, and crowdsourced data collection. Augmentation methods offer data generation to improve model performance. 3D printing allows for the creation of realistic replicas of landmines for safe experimentation. Crowdsourcing uses collective efforts to collect real-world data from conflict zones. Through the joint efforts of researchers, technology developers and humanitarian organizations, these approaches offer promising ways to improve landmine detection capabilities. The use of these approaches can address the data gap and ensure safe data collection.

Manyetik Anomali Yöntemi Kullanılarak Düşük Maliyetli Mayın Dedektörü Tasarımı

Buried mines pose great dangers to humans and animals around the world, which means thousands of people die each year from buried mines. Detecting and destroying these mines without harming people is an important issue. Today, these landmines are detected using different methods such as Ground Effect Radar, Electromagnetic induction, Infrared and Nuclear Quadrupole Resonance and active sensors are generally used in most of these methods. Although active sensor-based landmine detectors are often used for performance reasons, they can cause unintentional landmine explosions because they operate with transmitted and reflected signals. On the other hand, there is no such obstacle that can perform better in passive sensor-based landmine detectors depending on the design criteria. Therefore, in this study, a prototype design including passive sensor with magnetic anomaly method has been developed and shown. For the performance analysis of this detector, real landmines are used and the designed system is tested with different distance values in different soil types. The results show that the prototype produced successfully detects different types of landmines, is assertive in its lightness and only 1750 grams with its battery, providing sensitivity as well as advantages such as ease of use and low cost. It also shows the feature of being the first handheld landmine detector based on magnetic anomaly.

AACP 3.0

AACP 3.0

Electrical Impedance Tomographic Imaging of Buried Landmines

A prototype confirmation landmine detector, based on electrical impedancetomography (EIT), which can operate under realistic environmental conditions, has been developed. Laboratory and field experiments demonstrated that it is possible to reliably reconstruct, on the scale of the electrode spacing (ES) (in width and depth), conductivity perturbations due to a shallow buried antitank mine or a similar object in a variety of soils (black earth, clay, sand) down to depths equal to the dimensions of the object (1-1.5 ES, equivalent to 14-21 cm for a 64-electrode 1 m times 1 m array). These represent the first EIT images of real landmines computed from measured data. Occasional problems were encountered with the electrical contact in very dry soils, with excessive insertion pressure being required for reliable electrode contact. However, poor contacts could be detected, and the offending probe was either reinserted or compensation was applied. A matched filter detection algorithm based on a replica of the object of interest was developed and shown to effectively reduce the false alarm rate of the detector. EIT is especially suited for wet lands and underwater, where other mine detectors perform poorly. Experiments in a water-and sediment-filled tank have demonstrated that detection of minelike objects in such an environment with a submerged array is feasible. These experiments represent the first EIT measurements of targets using an electrode array submerged underwater. EIT may also have an application in locating intact mines in the berms formed by mine-clearing equipment. The EIT sensor head could be made cheaply enough to be disposable and remotely inserted to improve safety

14N-NQR based device for detection of explosives in landmines

A short survey of technologies proposed and implemented so far for landmine detection is given. A laboratory prototype device intended for RDX and HMX explosive detection in landmines by means of 14N-NQR (nuclear quadrupole resonance) spectroscopy is described. This NQR based landmine detector is essentially a fully automated and computer controlled FT-NQR spectrometer equipped with a planar r.f. measure coil. The temperature dependencies of NQR frequencies ν Q ( T) in RDX and HMX in the temperature range of practical interest for explosive detection were measured. Final testing of the device has been carried out on samples of sodium nitrite (NaNO 2) which had to be used to simulate real landmines in order to obey the safety regulations. Experimentally optimized SORC (strong off-resonance comb) multi-pulse sequence were used for optimum NQR signal. The time of detection was 30 and 90 s for sodium nitrite simulants buried in depths of 7 and 10 cm under ground, respectively. The sensitivity of detection was limited mostly by the external r.f. interferences, both man-made and naturally occurring, which enter the NQR detection system through the unshielded measure coil. Possible means to circumvent these limitations are also discussed.

Landmine Detection and Discrimination Using High-Pressure Waterjets

Methods of locating and identifying buried landmines using high-pressure waterjets were investigated. Methods were based on the sound produced when the waterjet strikes a buried object. Three classification techniques were studied, based on temporal, spectral, and a combination of temporal and spectral approaches using weighted density distribution functions, a maximum likelihood approach, and hidden Markov models, respectively. Methods were tested with laboratory data from low-metal content simulants and with field data from inert real landmines. Results show that the sound made when the waterjet hit a buried object could be classified with a 90% detection rate and an 18% false alarm rate. In a blind field test using 3 types of harmless objects and 7 types of landmines, buried objects could be accurately classified as harmful or harmless 60%-90% of the time. High-pressure waterjets may serve as a useful companion to conventional detection and classification methods.

Enhanced Auditory Displays for Discriminating Landmines from Clutter Using Electromagnetic Induction Sensors

Landmine detection is primarily performed using electromagnetic induction (EMI) sensors. These sensors detect the presence of metal and convey the information to the sensor operators via an audio signal. Reduction of false alarms from objects that contain metal but are not landmines, i.e. discrimination, is a challenging problem. Recent work on automated algorithms has shown promise towards reducing false alarm rates of EMI sensors. In this study, the audio signal was modified to encode the presence of metal as well as information regarding mine/non-mine belief in order to determine whether the additional information enabled operators to better discriminate mines from clutter. Using data collected from real landmines, we experimentally investigated which perceptual dimensions most effectively convey different aspects of the information contained in the sensor response to a listener. Results indicated that the presence of metal (detection) could be coded in the fundamental frequency of the audio signal, and that mine/non-mine belief (discrimination), determined using an automated algorithm, could be coded in a separate audio dimension. Operators performed better with this audio coding scheme than one where only metal content information was presented via the fundamental frequency of the audio signal.