In herbicide-selectivity experiments, response can be measured by visual inspection, stand counts, plant mortality, and biomass. Some response types are relative to nontreated control. We developed a nondestructive method by analyzing digital color images to quantify color changes in leaves caused by herbicides. The range of color components of green and nongreen parts of the plants and soil in Hue, Saturation, and Brightness (HSB) color space were used for segmentation. The canopy color changes of barley, winter wheat, red fescue, and brome fescue caused by doses of a glyphosate and diflufenican mixture, cycloxydim, diquat dibromide, and fluazifop-p-butyl were described with a log-logistic dose–response model, and the relationship between visual inspection and image analysis was calculated at the effective doses that cause 50% and 90% response (ED50 and ED90, respectively). The ranges of HSB components for the green and nongreen parts of the plants and soil were different. The relative potencies were not significantly different from one, indicating that visual and image analysis estimations were about the same. The comparison results suggest that image analysis can be used to assess color changes of plants in response to some herbicides and may have the potential to provide an objective measurement of symptoms.

Visual inspection remains the most frequently applied method for detecting treatment effects in single-case designs. The advantages and limitations of visual inference are here discussed in relation to other procedures for assessing intervention effectiveness. The first part of the paper reviews previous research on visual analysis, paying special attention to the validation of visual analysts' decisions, inter-judge agreement, and false alarm and omission rates. The most relevant factors affecting visual inspection (i.e., effect size, autocorrelation, data variability, and analysts' expertise) are highlighted and incorporated into an empirical simulation study with the aim of providing further evidence about the reliability of visual analysis. Our results concur with previous studies that have reported the relationship between serial dependence and increased Type I rates. Participants with greater experience appeared to be more conservative and used more consistent criteria when assessing graphed data. Nonetheless, the decisions made by both professionals and students did not match sufficiently the simulated data features, and we also found low intra-judge agreement, thus suggesting that visual inspection should be complemented by other methods when assessing treatment effectiveness.

For some companies, visual inspection has become an essential step when seeking to improve the quality of their products. The aim of this control is to be sure of the perceived quality of the product, which often goes well beyond the quality expected by the customer. For this type of control, the controller should be able to detect any anomaly on a product, characterize this anomaly, and then evaluate it in order to decide if the product should be accepted or rejected. This paper describes how this characterization can be carried out and, more specifically, how to measure the impact of the local environment of an anomaly on the perceived quality of the product.

This paper presents and compares different approaches currently used to assess surface anomalies identified on a product. The common point between these methods is that they are based on a document presented in the form of a table, which is to help the inspector to assess the anomaly detected in a repeatable and reproducible way. We will present three types of table: criteria/level table, tree-like presentation table and an indexed table. As each of these tables presents certain limits when applied to the inspection of a product surface, we will describe the table proposed in order to help the inspector determine the intensity to attribute to the identified anomaly.