NFNik Fadzly

Nik Fadzly

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Nik Fadzly has a total of 135 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Comparative evaluation of artificial intelligence-integrated and conventional camera traps for small mammal monitoring

    by Nurkamilia Azhar, Nik Fadzly

    Context: Conventional camera traps generate vast amounts of non-target images that makes edge artificial intelligence (AI) cameras a promising solution for real-time species detection. However, the real-world effectiveness of these systems depends heavily on how models generalise from training to unpredictable field conditions. Aims: We test whether within-distribution validation accuracy predicts field deployment performance for edge-deployed object detection models by examining how training-dataset background composition affects field generalisation for small mammal detection, and compare AI-integrated and conventional camera-trap detection outcomes under matched field conditions. Methods: We conducted a field-based comparison between an AI-integrated camera system (MaixCAM-Pro running YOLOv11n) and a conventional trail camera (Hawkray HC-900A) for small mammal monitoring. Three object-detection models were evaluated based on training datasets with different background compositions: heterogeneous varied backgrounds (Model A), uniform white backgrounds (Model B) and a combined dataset integrating both (Model C). Key results: Although Model B achieved the highest validation accuracy during training, field deployment revealed a severe performance gap. Model C demonstrated the strongest real-world performance, with 88.9% of MaixCAM-Pro captures correctly containing squirrels and a false-positive rate of 11.1%, markedly lower than Model A’s 98.4%. Conversely, Model A performed poorly in the field due to extensive background-related misclassifications. When the AI-integrated camera is equipped with Model C, it significantly reduces false triggers compared to the conventional camera while maintaining reliable small mammal (squirrel) detection. Conclusions: Training-dataset background composition substantially affects field detection performance. The within-distribution validation accuracy does not reliably predict field deployment success. Balanced datasets incorporating both controlled and varied backgrounds are essential to improve model generalisation. Implications: Wildlife AI applications must validate models under field conditions matching the deployment context rather than relying solely on training metrics. Furthermore, the model’s precision profile must align with the downstream ecological inference framework (e.g. occupancy modelling) to ensure data integrity.

  • Open Access

    Optimising nocturnal mammal surveys in tropical forests: when red light is the key

    by Priscillia Miard, Ummiatul Zainalabidin, Ulmar Grafe, Salwa Khalid, Nik Fadzly Nik Rosely, Henry Bernard, Nadine Ruppert


  • Open Access

    A deep learning framework for bone fragment classification in owl pellets using YOLOv12

    by Nik Fadzly, Lay Wai Kean, Siti Nuramaliati Prijono, Rini Rachmatika, Siti Zulaika, Mohd Nasir, Hasber Salim

    Non-invasive monitoring of small mammal populations is critical for both biodiversity conservation and integrated pest management, particularly in agroecosystems. Barn owl (Tyto alba) pellet analysis has long served as a valuable tool for inferring prey abundance, yet conventional bone classification is labour-intensive and requires specialized expertise. Here, we introduce a deep learning framework that automates the detection and classification of rodent bone fragments from owl pellets using the YOLOv12 object detection architecture. A dataset comprising 978 annotated images, encompassing skull, femur, mandible, and pubis bones, was used to train and validate the model, achieving high detection performance (precision = 0.90, recall = 0.90, mAP@0.5 = 0.984, F1-score = 0.97). The model demonstrated strong generalization across samples from Malaysia and Indonesia. We further developed a Python-based inference script to estimate rodent abundance using skull and paired bone counts. This AI-assisted workflow reduces human error, increases processing throughput, and enables scalable rodent monitoring. By enhancing ecological inference from pellet studies, our approach supports timely biodiversity assessments and pest surveillance strategies across diverse landscapes.


  • Open Access

    Efficacy of Several Types of Pest Bird Deterrents and General Trend of Pest Birds at an Industrial Factory

    by Imran Mohd Hornain, Nik Fadzly N Rosely

    Controlling pest birds is a complex problem, especially for large areas, compared to individual homes. This study presents a new perspective on pest birds plaguing a large semiconductor factory. We evaluated the efficacy of nine bird deterrent methods: alarm and distress calls broadcasted from portable speaker, sound frequencies ranged from (i) 24.5 kHz–45.5 kHz; and (ii) 13.5 kHz–45.5 kHz together with flashing lights emitted from sonic device, methyl anthranilate (MA), moving and static bird predator models, reflective compact discs, high and low visibility reflective tapes in deterring Barn Swallow (Hirundo rustica), Pacific Swallow (Hirundo tahitica) and Asian Glossy Starling (Aplonis panayensis) from perching on targeted spots. Monthly counts of pest birds roosting at study site were also conducted. Our results showed that alarm and distress calls broadcasted from portable speaker, sound frequencies ranged from 13.5 kHz–45.5 kHz together with flashing lights emitted from sonic device, moving and static bird predator model, reflective compact discs, high and low visibility reflective tapes were significantly effective in deterring pest birds from targeted spots. The pest bird population reached its peak (35,063) in January 2021, while the least (3,685) was recorded in May 2021. The effectiveness of pest bird deterrents might be influenced by the quantity and method of deployment.


  • Open Access

    Evaluating drones as bird deterrents in industrial environments: multirotor vs fixed-wing efficacy

    by Imran Mohd Hornain, Nik Fadzly N Rosely

    Unmanned aerial vehicles (UAVs) or drones have been proposed as deterrent tools to mitigate pest birds’ problems. Many studies have been conducted to evaluate the efficacy of drones, mainly to protect crops, fishponds and airports. Little information can be acquired on using drones in industrial areas. In this study, two types of drones, categorized as multirotor drones and fixed-wing drones, were used to evaluate their efficacy in reducing pest birds, Asian glossy starling (Aplonis panayensis) flocks in one of the semiconductor factories in Kulim Hi-tech Park, Kedah, Malaysia during dusk. Each drone was evaluated during its five minutes of operation time and five minutes after landing. Control data were also taken to compare drone treatment days with no drone treatment days. Our result shows a significant difference between multirotor drone treatment and control treatment but not between fixed-wing drone treatment and control treatment due to different altitudes applied, ambient light intensity and size of flight path covered. We suggest implementing biomimetic design into drones and applying other conventional ground deterrents to prolong the residual effect of post-treatment.


  • Open Access

    Proportions of Green Area and Tree Health on University Campus: The Impact of Pavement Presence

    by Nur Ain Aiman Abd Rahim, Rahmad Zakaria, Asyraf Mansor, Mohd Ashraf Mohamad Ismail, Nik Fadzly N Rosely

    Pertanika Journal information and guidelines

  • Open Access

    Group densities of endangered small apes (Hylobatidae) in two adjacent forest reserves in Merapoh, Pahang, Malaysia

    by Adilah Suhailin Kamaruzaman, Nurul Iza Adrina Mohd Rameli, Susan Lappan, Thad Quincy Bartlett, Nik Rosely Nik Fadzly, Mohd Sah Shahrul Anuar, Nadine Ruppert

    Small ape habitat is rapidly declining due to anthropogenic activities but the current population status of this endangered primate family in Malaysia remains unknown. Group densities of Hylobates lar and Symphalangus syndactylus in two adjacent forest reserves across the Sungai Yu Ecological Corridor (SYEC) in Merapoh, a critical connectivity area of the Central Forest Spine, were assessed. Vegetation assessment and satellite imagery were used to identify habitat characteristics and fixed-point active acoustic triangulation at six listening areas was conducted to estimate small ape group densities. Small ape habitat quality was high in the forested areas of the SYEC. The mean group density of H. lar across these six areas was 3.55 ± 0.9 groups km-² while the mean group density of S. syndactylus was 2.75 ± 1.0 groups km-². The mean group densities of small apes at SYEC were moderately high, compared with densities at other sites in the region, which suggests that the forests here constitute good habitat for both species, despite some observed anthropogenic disturbances. Both species occurred in all listening areas. A nationwide population census for small apes and regular monitoring to inform conservation planning are recommended. Further improvement to connectivity across the SYEC by installing artificial canopy bridges for arboreal animals is important to support the movement of small apes across habitat fragments in Merapoh.


  • Open Access

    Development of species recognition models using Google teachable machine on shorebirds and waterbirds

    by Jenny Jenn Ney Wong, Nik Fadzly

    Species identification is an essential ability in every conservation initiative. An efficient and robust computer vision method was attested with an available online tool with Google's Teachable Machine. This pilot study on developing a species recognition app was to create and evaluate the usability and accuracy of using Teachable Machine for species identification at Teluk Air Tawar, Kuala Muda (TAT-KM), Malaysia. The accuracy of the created models was evaluated and compared with training images based on the web-mining (Google Images Repository) compared to actual photos taken at the same site. Model A (Google Image) had an average accuracy of 55.30%, while Model B (actual photos) was 99.42%. Regarding success rate at accuracy over 77%, 27 out of 49 test images (55.10%) were reported in Model A, while Model B had a 100% success rate. This approach can replace traditional methods of bird species recognition to handle large amounts of data.


  • Evaluation of the pathogenicity of endophytic fungi isolated from spines of rattan (Calamus castaneus) against other plant hosts

    by Nurul Farizah Azuddin, Masratul Hawa Mohd, Nik Fadzly Nik Rosely, Asyraf Mansor, Latiffah Zakaria

    Aims: In this study, we sought to determine the pathogenic ability of endophytic fungi recovered from the spines of Calamus castaneus, a common rattan palm growing in the forests of Peninsula Malaysia. Methods and Results: Ten endophytic fungal isolates were selected for pathogenicity tests: Colletotrichum boninense, Colletotrichum fructicola, Colletotrichum cliviae, Diaporthe hongkongensis, Diaporthe arengae, Diaporthe cf. nobilis, Neopestalotiopsis saprophytica, Neopestalotiopsis formicarum, Fusarium solani and Fusarium oxysporum. These endophytes were tested against leaves of bertam (Eugeissona sp.), oil palm (Elaeis guineensis) and mango (Mangiferae indica) and the fruits of chilli (Capsicum annum), tomato (Solanum lycopersicum) and banana (Musa acuminata). The fungal isolates showed infectivity against bertam, oil palm and mango leaves with degrees of virulence ranging from low to moderate, whereas infectivity against chilli, tomato and banana ranged from low to very high. Conclusions: Fungal endophytes isolated from the spines of C. castaneus are pathogenic to different crop plants with differing degrees of virulence or aggressiveness. Significance and Impact of the Study: Spines of C. castaneus can harbour fungal pathogens of a number of different crops as endophytes. The ability of the fungal endophytes to colonize and infect different crops demonstrate their importance towards agricultural crops. There is a possibility the endophytes behave as latent pathogen. When conditions become favourable, the fungal endophytes transform to pathogenic form and potentially infect other plants.