Automatic exposure algorithms for digital photography

Bernacki, Jarosław

doi:10.1007/s11042-019-08318-1

Cited by 16 publications

(8 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Color balance is used to manipulate any unwanted color that dominates an image by estimating the illumination and applying correction to the image [ 36 ]. Finally, exposure adjustment is focused on controlling the light of on an image via two parameters: the exposure time and the light sensitivity of the image [ 37 ]. Enhanced images are presented in Figure 4 .…”

Section: Experimental Results and Analysismentioning

confidence: 99%

Non-Intrusive Fish Weight Estimation in Turbid Water Using Deep Learning and Regression Models

Tengtrairat

Woo

Parathai

et al. 2022

Sensors

View full text Add to dashboard Cite

Underwater fish monitoring is the one of the most challenging problems for efficiently feeding and harvesting fish, while still being environmentally friendly. The proposed 2D computer vision method is aimed at non-intrusively estimating the weight of Tilapia fish in turbid water environments. Additionally, the proposed method avoids the issue of using high-cost stereo cameras and instead uses only a low-cost video camera to observe the underwater life through a single channel recording. An in-house curated Tilapia-image dataset and Tilapia-file dataset with various ages of Tilapia are used. The proposed method consists of a Tilapia detection step and Tilapia weight-estimation step. A Mask Recurrent-Convolutional Neural Network model is first trained for detecting and extracting the image dimensions (i.e., in terms of image pixels) of the fish. Secondly, is the Tilapia weight-estimation step, wherein the proposed method estimates the depth of the fish in the tanks and then converts the Tilapia’s extracted image dimensions from pixels to centimeters. Subsequently, the Tilapia’s weight is estimated by a trained model based on regression learning. Linear regression, random forest regression, and support vector regression have been developed to determine the best models for weight estimation. The achieved experimental results have demonstrated that the proposed method yields a Mean Absolute Error of 42.54 g, R2 of 0.70, and an average weight error of 30.30 (±23.09) grams in a turbid water environment, respectively, which show the practicality of the proposed framework.

show abstract

Section: Experimental Results and Analysismentioning

confidence: 99%

Non-Intrusive Fish Weight Estimation in Turbid Water Using Deep Learning and Regression Models

Tengtrairat

Woo

Parathai

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…To this end, we vary the intensity of the stripe pattern by further introducing the term I e , which is to model the intensity of environment as I e = ratio * I p . We consider the ratio ranging in [0, 1.4] with intensity normalization based on the auto-ISO property [44]. The results are shown in Fig.…”

Section: Discussionmentioning

confidence: 99%

“…As we aim to model the stripe pattern based on the existing image, the additional illumination intensity (for capturing new images) does not need to be considered. Therefore, we set I e = 0 with auto-ISO property through intensity normalization [44] with I p = 100.…”

Section: Simulation Evaluationmentioning

confidence: 99%

Light Can Hack Your Face! Black-box Backdoor Attack on Face Recognition Systems

Li,

Wang,

Xie

et al. 2020

Preprint

View full text Add to dashboard Cite

Deep neural networks (DNN) have shown great success in many computer vision applications. However, they are also known to be susceptible to backdoor attacks, where the misclassification behaviors are hidden in a normal DNN model, but can be triggered by a specific pattern. When conducting backdoor attacks, most of the existing approaches assume that the targeted DNN is always available, and an attacker can always inject a specific pattern to the training data to further fine-tune the DNN model. As such, a testing input with the specific pattern will be misclassified by the targeted DNN model. However, in practice, such attack may not be feasible as the DNN model is encrypted and only available to the secure enclave.In this paper, we propose a novel black-box backdoor attack technique on face recognition systems, which can be conducted without the knowledge of the targeted DNN model. To be specific, motivated by the recent advance of DNN interpretation that DNN explores common knowledge such as edge and color, we propose a backdoor attack with a novel color stripe pattern trigger, which can be generated by modulating LED in a specialized waveform. We also use an evolutionary computing strategy to optimize the waveform for backdoor attack. Our backdoor attack can be conducted in a very mild condition: 1) the adversary cannot manipulate the input in an unnatural way (e.g., injecting adversarial noise); 2) the adversary cannot access the training database; 3) the adversary has no knowledge of the training model as well as the training set used by the victim party. We show that the backdoor trigger can be quite effective, where the attack success rate can be up to 88% based on our simulation study and up to 40% based on our physical-domain study by considering the task of face recognition and verification based on at most three-time attempts during authentication. Finally, we evaluate several state-of-the-art potential defenses towards backdoor attacks, and find that our attack can still be effective. We highlight that our study revealed a new physical backdoor attack, which calls for the attention of the security issue of the existing face recognition/verification techniques.

show abstract

“…Hal tersebut harus diimbangi dengan kemampuan seseorang yang mengoperasikan kamera tersebut harus memahami dasar fotografi berupa perhitungan pencahayaan. Faktor terpenting dalam fotografi adalah pengaturan exposure yang maksimal [1]. Eksposur dapat dipahami sebagai kombinasi optimal dari nilai aperture ( f ), kecepatan rana ( t ) dan sensitivitas (ISO) karenanya pengaturan eksposur yang benar bukanlah tugas yang mudah.…”

Section: Pendahuluanunclassified

“…Alat light meter sendiri memiliki perhitungan berdasarkan nilai intensitas cahaya (lux) yang artinya cahaya sebagai objek utama pengolahan komposisi didalam alat light meter. Penelitian tentang perhitungan komposisi eksposur yang pernah dilakukan [1], penelitian tersebut membahas tentang algoritma eksposur otomatis (AE) dalam kamera digital. Hasil dari penelitian tersebut algoritma baru mampu mendeteksi nilai overexposure lebih akurat sehingga gambar yang dihasilkan dapat lebih seimbang.…”

Section: Pendahuluanunclassified

Prototipe Light Meter Fotografi Studio Menggunakan Mikrokontroler ATMega328 Berbasis Sensor Cahaya dan Warna

Gunawan

Fauziah

Hayati

2021

mib

View full text Add to dashboard Cite

Accuracy in calculating exposure in photographic techniques is the key to being able to produce good quality photos. The light calculation includes three elements (Triangel Exposure), namely shutter speed (shutter speed / t), ISO sensitivity, and diaphragm (aperture / f). A light meter is a tool for calculating lighting in the world of photography to accurately calculate the light falling in or around the object to be photographed. The light meter has a working principle like a lux meter calculates the value of light intensity in a place. However, on a light meter for photography, there is a calculation of a triangle exposure in it. This light meter prototype uses an ATMega328 microcontroller on the Arduino Nano V3 with the help of a TSL2561 light sensor and a TCS34725 RGB color sensor. Tests were carried out by comparing the output value using an original light meter, namely Sekonik Flashmate L-308X with the continuous light shooting method. The results of the test using the original tool with the ATMega328 microcontroller produced values above 90%.

show abstract

Automatic exposure algorithms for digital photography

Cited by 16 publications

References 9 publications

Non-Intrusive Fish Weight Estimation in Turbid Water Using Deep Learning and Regression Models

Non-Intrusive Fish Weight Estimation in Turbid Water Using Deep Learning and Regression Models

Light Can Hack Your Face! Black-box Backdoor Attack on Face Recognition Systems

Prototipe Light Meter Fotografi Studio Menggunakan Mikrokontroler ATMega328 Berbasis Sensor Cahaya dan Warna

Contact Info

Product

Resources

About