A New Facial Expression Recognition Technique using 2-D DCT and Neural Networks Based Decision Tree
Bibliographic record
Abstract
Human facial expression recognition (FER) has attracted much attention in recent years because of its importance in realizing highly intelligent human-machine interfaces. In this paper, we propose a new FER technique that utilizes the 2-D DCT of full size facial images and a decision tree with feedforward neural network (NN) based nodes. The first NN-based node of the decision tree is designated to separate one group of facial expressions with members "smile" and "surprise" from another group that contains "anger" and "sadness". This node can reduce the confusion between the category members of the two groups. Two NN-based nodes that follow the first node are established for each group to separate their two members. As a result, the original recognition problem with four categories is divided into three subproblems, each having only two members to distinguish. This work is the first trial to use NN, decision tree and 2-D DCT simultaneously within a single recognition task. To demonstrate the capability of the proposed recognition technique, we use two databases, including a recently constructed one, which contain 2-D front face images of 60 men and 60 women, respectively. Experimental results reveal that the new technique outperforms, on the whole, the simple vector matching and K-means based vector matching techniques and two recently developed methods using fixed-size and constructive neural networks. The mean recognition rates of the new technique have been found as high as 97.5% and 93.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".